Borehole geology and sub-surface petrochemistry of the Domes area, Olkaria

geothermal field, Kenya, in relation to well OW-922.

Victor Omondi Otieno

Faculty of Earth Science

University of Iceland

2016

Borehole geology and sub-surface petrochemistry of the Domes area, Olkaria geothermal field, Kenya, in relation to well

OW-922.

Victor Omondi Otieno

60 ECTS thesis submitted in partial fulfillment of a Magister Scientiarum degree in Geology

Advisor(s)

Anette K. Mortensen

Björn S. Harðarson

Faculty Representative

Guðmundur H. Guðfinnsson Sæmundur A. Halldórsson

Examiner Dr. Bjarni Gautason

Faculty of Earth Science School of Engineering and Natural Sciences

University of Iceland Reykjavik, May 2016

Borehole geology and sub-surface petrochemistry of the Domes area, Olkaria geothermal

field, Kenya, in relation to well OW-922.

60 ECTS thesis submitted in partial fulfillment of a Magister Scientiarum degree in

Geology

Copyright © 2016 Victor Omondi Otieno

All rights reserved

Faculty of Earth Science

School of Engineering and Natural Sciences

University of Iceland

Sturlugata 7

101, Reykjavik

Iceland

Telephone: 525 4000

Bibliographic information:

Victor Omondi Otieno, 2016,

Borehole geology and sub-surface petrochemistry of the Domes area, Olkaria geothermal

field, Kenya, in relation to well OW-922.

Master’s thesis, Faculty of Earth Science, University of Iceland, pp. 144.

Printing: Háskólaprent, Fálkagata 2, 107 Reykjavík, Iceland, May 2016

Abstract

Olkaria is a high temperature geothermal system located within the central sector of the

Kenya Rift System, and associated with a region of Quaternary volcanism. Recharge into

the system is exclusively by meteoric water, derived primarily from the high-altitude rift

escarpments. Tectonic structures play without a doubt a pivotal role in enhancing fluid flow

within the geothermal system. For this study, data and samples from four wells, OW-905A,

OW-910, OW-917 and OW-922, have been used. Wells OW-905A, OW-910 and OW-917

were drilled within and along the proposed caldera structure. Well OW-922 is a step-out

vertical well, drilled to the east of the proposed caldera structure. This study largely focuses

on well OW-922, while, the other three wells have been included for comparison purposes.

The study presents a comprehensive look at the borehole geology and hydrothermal

mineralisation, sub-surface petrochemistry and microprobe analysis. Lithologies intersected

by the study well include pyroclastics, tuff, rhyolite, basalt, trachydacite, basaltic

trachyandesite, trachyandesite and trachyte. No intrusions are found in this well.

Hydrothermal products in the study well occur both as replacement of primary components

or glassy matrix and as open space filling in veins, fractures and vesicles. Evidence of a

significant cooling process (up to over 110oC) in the geothermal system around the study

well is provided by the difference between measured formation temperature (127oC) and

inferred hydrothermal alteration temperature (epidote 240oC). Cases of calcite overprinting

epidote further points towards a cooling process.

Five feed zones have been deduced from the well and all are categorised as small. Generally,

the well has low permeability and could not sustain discharge after 49 days of heat-up.

Observations from temperature distribution across the field indicates that the study well is

located outside the Olkaria Domes heat source, and consequently, outside the main Olkaria

volcanic zone. Whole-rock chemistry displays a range of compositions of the samples of the

study well, from basalt to trachyte or rhyolite, with the prevalence of the rocks being the

highly evolved derivatives. Major oxides and trace element systematics are dominated by

crystal fractionation, even though other differentiation processes cannot be ruled out.

Evidence of a common differentiation mechanism for the surface and sub-surface rocks

seems likely on the basis of comparison between the two groups of samples. Microprobe

analyses have been carried out on epidote and chlorite in order to assess the chemical

composition and the compositional range of the two minerals. It is found that most of the

epidote are aluminium-rich, whereas, the chlorites are iron-rich varieties. Limited number of

epidote and chlorite has however, impeded possible correlation of the stratigraphy in the four

wells based on major and minor element contents.

Útdráttur

Olkaria er háhitasvæði innan miðhluta gliðnunarbeltis Keníu (Kenya Rift System) og tengist

eldgosavirkni á kvartertíma. Vatn sem streymir inn í jarðhitakerfið er að uppruna regnvatn,

sem fellur aðallega á hálendið meðfram sprungudalnum. Tektóník er vafalaust mikilvægur

þáttur í að auka vatnsstreymi innan jarðhitakerfisins. Gögn og sýni frá fjórum borholum,

OW-905A, OW-910, OW-917 og OW-922, hafa verið notuð í þessari rannsókn. Holur OW-

905A, OW-910 and OW-917 voru boraðar innan og á börmum ætlaðrar öskju. OW-922 er

lóðrétt hola, boruð utan svæðis með fyrri holum, austur af öskjunni. Í þessari rannsókn er

augum aðallega beint að holu OW-922, en hinar holurnar hafðar með til samanburðar.

Ýtarlega er litið á borholujarðfræði, myndun jarðhitasteinda og efnafræði

neðanjarðarmyndana og efnasamsetning steinda var greind með örgreini. Meðal myndana

sem fundust í holunni eru gjóska, túff, rhýólít, basalt, trakýdasít, basaltískt trakýandesít,

trakýandesít og trakýt. Ekki sáust merki um innskot í þessari borholu. Jarðhitaútfellingar

hafa komið í stað upprunalegra steinda og gosglers og sest í opið rými í sprungum, æðum og

blöðrum. Merki sjást um að umtalsverð kæling (um eða yfir 110°C) hafi orðið í

jarðhitakerfinu kringum borholuna út frá muninum á mældu hitastigi (127°C) og áætluðu

ummyndunarhitastigi (epidót, 240°C). Dæmi um að kalsít hafi vaxið yfir epidót benda einnig

til að kæling hafi átt sér stað.

Merki sjást um fimm lektarsvæði í holunni, sem öll flokkast sem minni háttar. Almennt hefur

borholan lága lekt og hélst ekki í blæstri eftir 49 daga hitun. Samanburður á hita í borholum

á svæðinu bendir til að rannsóknarsvæðið sé utan varmagjafa Domes-svæðisins í Olkaria og

þar af leiðandi utan aðal eldgosasvæðis Olkaria. Efnasamsetning bergsýna úr holunni er

breytileg frá basalti til trakýts eða rhýólíts, en þróað berg er ríkjandi. Breytileiki bæði

aðalefna og snefilefna bendir til að magn þeirra stjórnist mest af kristaldiffrun, þó ekki sé

hægt að útiloka önnur þróunarferli. Samanburður á yfirborðs- og borholusýnum bendir til

sams konar þróunarferla fyrir báða hópa sýna. Samsetning epidóts og klóríts var greind með

örgreini til að fá upplýsingar um efnasamsetningu og breytileika í samsetningu þessara

steinda. Epidót reynist álríkt, en klórít er af járnríkri gerð. Þar sem fá sýni með epidóti og

klóríti fundust er hins vegar erfitt að nota breytileika í magni aðalefna og snefilefna í þeim

til að tengja saman jarðlög í borholunum fjórum.

Dedication

This thesis is dedicated to my beloved wife, Vivian and my lovely daughter, Tasha.

vii

Table of Contents

List of Figures ..................................................................................................................... xi

List of Tables ...................................................................................................................... xv

Abbreviations ................................................................................................................... xvii

Acknowledgements ........................................................................................................... xix

1 Introduction ..................................................................................................................... 1 1.1 General overview .................................................................................................... 1 1.2 Overview of well OW-922 ...................................................................................... 3

1.3 Goals of the study .................................................................................................... 4 1.4 Thesis layout............................................................................................................ 5

2 Outline of Geology .......................................................................................................... 7 2.1 Regional geology ..................................................................................................... 7

2.2 Geology of Olkaria Volcanic Complex ................................................................... 9

2.2.1 Surface geology ........................................................................................... 10 2.2.2 Sub-surface geology..................................................................................... 13

2.3 Structural geology and tectonics of OVC.............................................................. 15

2.4 Hydrogeological setting ........................................................................................ 17 2.5 Reservoir properties of Olkaria field ..................................................................... 19

2.6 Summary of drilling history .................................................................................. 20 2.6.1 Phase 1: Drilling of the Surface hole ........................................................... 21 2.6.2 Phase 2: Drilling of the Intermediate hole ................................................... 21

2.6.3 Phase 3: Drilling of Production hole ............................................................ 22 2.6.4 Phase 4: Drilling of Production section (Main hole) ................................... 22

3 Sampling and Analytical Methods .............................................................................. 23 3.1 Sampling ................................................................................................................ 23 3.2 Analytical methods ................................................................................................ 23

3.2.1 Binocular microscope analysis .................................................................... 23 3.2.2 Petrographic microscope analysis ................................................................ 24

3.2.3 X-ray diffraction analysis ............................................................................ 24 3.2.4 Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-

OES) analysis ............................................................................................... 24 3.2.5 Electron microprobe analysis ....................................................................... 25

4 Results ............................................................................................................................ 27 4.1 Lithology of well OW-922 .................................................................................... 27

4.1.1 Pyroclastics .................................................................................................. 27

4.1.2 Tuffs ............................................................................................................. 27 4.1.3 Rhyolite ........................................................................................................ 28 4.1.4 Basalt............................................................................................................ 28

viii

4.1.5 Basaltic trachyandesite ................................................................................ 28 4.1.6 Trachyandesite ............................................................................................. 28 4.1.7 Trachydacite ................................................................................................ 29 4.1.8 Trachyte ....................................................................................................... 29

4.2 Hydrothermal alteration ......................................................................................... 29 4.2.1 Distribution of hydrothermal alteration minerals ........................................ 30 4.2.2 Hydrothermal alteration zonation ................................................................ 39 4.2.3 Distribution of alteration mineral zones in the Domes area ........................ 40

4.3 Feed zones .............................................................................................................. 42

4.4 Alteration and measured formation temperatures .................................................. 46 4.5 Whole-rock chemistry ............................................................................................ 47

4.6 Major and trace element Compositions ................................................................. 47 4.6.1 Major elements ............................................................................................ 48 4.6.2 Trace elements ............................................................................................. 50

4.7 Classification of rock types .................................................................................... 52 4.7.1 TAS classification scheme ........................................................................... 53

4.7.2 Al2O3 vs. FeO classification scheme ........................................................... 54 4.8 Comparison of well OW-922 with surface and other sub-surface samples ........... 59

4.8.1 Comparison with surface samples ............................................................... 59 4.8.2 Comparison with other sub-surface samples ............................................... 60

4.9 Microprobe analysis ............................................................................................... 62

5 Discussion ....................................................................................................................... 71 5.1 Lithological units ................................................................................................... 71 5.2 Hydrothermal alteration ......................................................................................... 71

5.3 Permeability and comparison of temperatures ....................................................... 72 5.4 Rock composition .................................................................................................. 75 5.5 Hydrothermal alteration effects on rock chemistry ................................................ 75

5.5.1 Effects on major element geochemistry ...................................................... 75 5.5.2 Effects on trace element geochemistry ........................................................ 77

5.6 Magmatic differentiation processes ....................................................................... 78 5.6.1 Major elements ............................................................................................ 78 5.6.2 Trace elements ............................................................................................. 79

5.6.3 Assessing other differentiation mechanisms ............................................... 80

5.7 Comparison of surface and sub-surface samples ................................................... 81

5.8 Compositional variation of alteration minerals ...................................................... 82 5.8.1 Epidote ......................................................................................................... 82

5.8.2 Chlorite ........................................................................................................ 84

6 Conclusion and Recommendations .............................................................................. 85 6.1 Conclusion ............................................................................................................. 85 6.2 Recommendations .................................................................................................. 86

References ........................................................................................................................... 87

Appendix A: Lithological description of well OW-922 based on binocular and optical

petrographic analyses. ..................................................................................................... 93

Appendix B: XRD analysis for clay minerals in well OW-922. ........................................ 99

ix

Appendix C: ICP-OES analysis procedure ...................................................................... 103

Appendix D: EMP analysis for chlorite and epidote ........................................................ 105

xi

List of Figures

Figure 1: Map of the KRS showing the location of Olkaria Volcanic Complex and

other surrounding volcanic centres .................................................................... 1

Figure 2: The sub-fields of Olkaria geothermal field............................................................ 2

Figure 3: A section of the Olkaria Domes field showing location of the step-out well

OW-922 and other study wells OW-910, 917 and 905A. Also shown are

other drilled wells. The black triangles delineate the outline of the

proposed ring structure. ..................................................................................... 4

Figure 4: DEM of the EARS. Black lines represent main rift faults, whereas, white

surfaces are lakes ............................................................................................... 7

Figure 5: Map of the EARS showing elevations higher than 1200 m (black areas),

evidencing the main Ethiopian and Kenyan Plateaux (dashed lines). The

black areas mark the Cenozoic rocks of the Rift ................................................ 8

Figure 6: Geological map of the Olkaria Volcanic Complex and the surrounding area

.......................................................................................................................... 10

Figure 7: Surface formations and stratigraphic column of Olkaria Volcanic Complex.

Also shown are a series of eruptive peripheral mafic and trachyte

formations ......................................................................................................... 12

Figure 8: Geographical extents and stratigraphic units of the Olkaria eruptive centres.

The Ol-Njorowa Pantellerite Formation (O1) is not shown ............................. 13

Figure 9: Sub-surface lithotypes of the Olkaria Volcanic Complex ................................... 14

Figure 10: Structural map of the Olkaria Volcanic Complex displaying the orientation

of various fault patterns .................................................................................... 16

Figure 11: Piezometric map showing reservoir recharge areas for Olkaria and other

surrounding geothermal fields along the rift axis ............................................ 18

Figure 12: A vertical schematic sketch of the conceptual model of Olkaria geothermal

system. The suggested heat sources, up-flow zones and temperature cross-

section are shown. The green arrow is directed north ..................................... 19

Figure 13: Drilling progress curve for well OW-922 ......................................................... 21

Figure 14: Lithology, location of feed zones, distribution of alteration minerals and

alteration zones of well OW-922 ...................................................................... 32

xii

Figure 15: X-ray diffractogram of air dried, glycolated and heated sample of smectite.

.......................................................................................................................... 36

Figure 16: Cross polarized image of mixed-layer clay depicting high interference

colours at a depth of 2782 m. ........................................................................... 37

Figure 17: A plane polarized image of fibrous chlorite filling a vesicle at 1826 m. .......... 38

Figure 18: X-ray diffractogram of illite and unstable chlorite showing obscured peak

of 14.18 Å for the three treatments and 7.15 Å peak for air dried and glycol

solvation. The latter peak totally collapses upon heating. ............................... 39

Figure 19: Distribution of alteration mineral zones across a section of the Domes area,

along the traverse A-B. ..................................................................................... 41

Figure 20: Downhole pressure recovery logs for well OW-922. ........................................ 43

Figure 21: Feed zones in well OW-922 as deduced from temperature recovery logs,

drilling parameters and geological observations. ........................................... 45

Figure 22: Comparison of inferred alteration and measured formation temperatures

along with boiling point curve for well OW-922. ............................................. 47

Figure 23: Harker diagrams showing major element variations against SiO2 for

samples from well OW-922 .............................................................................. 49

Figure 24: Variation diagrams for a range of trace elements plotted against the index

of differentiation, SiO2 for samples from well OW-922 ................................... 51

Figure 25: TAS diagram showing the compositional range for sub-surface rocks from

well OW-922. .................................................................................................... 53

Figure 26: Al2O3 vs. FeO classification of sub-surface rocks from well OW-922. ............. 55

Figure 27: TAS classification scheme showing comparison of sub-surface samples

from wells OW-922, 905A, 910 and 917 and, surface samples from Olkaria

and neighboring minor eruption centres (represented by broken lines). ......... 59

Figure 28: A comparison for variation in concentration of selected major element

oxides for wells OW-922, 905A, 910 and 917 .................................................. 60

Figure 29: A comparison for variation in concentration of selected trace elements for

wells OW-922, 905A, 910 and 917 ................................................................... 61

Figure 30: BSE image of epidote from 864 m in well OW-910. Compositional

heterogeneity is clearly seen in this figure with Al-rich sections appearing

darker and Fe-rich sections lighter in color. Numerous compositional

zonation patterns are represented by Al-rich rims and Fe-rich cores. Also,

note the titanite inclusions in epidote. .............................................................. 65

xiii

Figure 31: BSE image of epidote filling a vesicle from 1560 m in well OW-905A.

Epidote is observed occuring as aggregates of small columnar cystals.

Quartz is also noted growing in the vesicle. Black areas in the lower right

hand corner, at the centre and within quartz are epoxy. .................................. 65

Figure 32: A WDS elemental map showing typical compositional zoning of epidote

from 864 m in well OW-910. Grayscale map (a) represent the original

composition, whereas, maps (b), (c) and (d) denote Fe, Sr and Al levels

respectively. ...................................................................................................... 67

Figure 33: A WDS elemental map showing Fe (b) and Al (d) distribution within a

single epidote grain from 2782 m in well OW-922. The darker and lighter

sections in the original compositional map (a) indicates Al and Fe-rich

regions, respectively. Map (c) shows Sr levels. ................................................ 68

Figure 34: BSE image of chlorite replacing plagioclase at 1870 m in well OW-917. ........ 70

Figure 35: A possible conceptual model based on outcome of the study, along the

traverse A-B. ..................................................................................................... 74

Figure 36: Variation diagrams showing the concentrations of Na2O, Al2O3, K2O and

MnO against SiO2. ............................................................................................ 76

Figure 37: Variation diagrams showing the concentrations of Rb, Sr, La, Ni, Ba and

Y against Zr. ..................................................................................................... 77

Figure 38: A plot of La vs. Zr. and Y vs. Zr. showing different magmatic differentiation

processes. The dotted and continuous lines indicates fractional

crystallisation and other petrogenetic processes, respectively. ....................... 80

Figure 39: Harker's variation diagrams between various pairs of incompatible trace

elements displaying fractional crystallisation (dotted line). The flexure

from the main trend (continuous line) represent other magma

differentiation processes. .................................................................................. 82

xiv

xv

List of Tables

Table 1: Utilization of geothermal energy for electric power generation in Olkaria. .......... 3

Table 2: Drilling phases and casing design for well OW-922. ........................................... 20

Table 3: Alteration of primary minerals and order of replacement in well OW-922 ......... 30

Table 4: Interpreted permeable zones based on geological observations. ......................... 46

Table 5: Major and trace element results for well OW-922 based on ICP-OES analysis.

.......................................................................................................................... 56

Table 6: Representative EMP analyses of epidote from wells OW-922, 910 and 905A. .... 64

Table 7: Range and average molar proportion of pistachite (Xps) component in

epidote............................................................................................................... 66

Table 8: Representative EMP analyses of chlorite for wells OW-922, 910, 917 and

905A. ................................................................................................................. 69

Table 9: Molar ratio of chlorite from wells OW-922 and 910. ........................................... 70

xvii

Abbreviations

AUs- Additional Units

BP- Before Present

BSE image- Backscattered Electron image

CKPP- Central Kenya Peralkaline Province

CT- Cellar Top

DEM- Digital Elevation Model

EARS- East African Rift System

EMP- Electron Microprobe analysis

GWDC- Greatwall Drilling Company Ltd.

HFSE- High Field Strength Elements

ICP-OES- Inductively Coupled Plasma-Optical Emission Spectrometry

ITEs- Incompatible Trace Elements

ÍSOR - Iceland Geosurvey

ka- Thousand years

KenGen- Kenya Electricity Generating Company Ltd.

KRS- Kenyan Rift Sytem

LILE- Large Ion Lithophile Elements

LREE- Light Rare Earth Elements

Ma- Million years

Masl- Metres above sea level

MER- Main Ethiopian Rift

MLC- Mixed Layer Clay

MWe – Mega Watt electrical

OD- Outside Diameter

OVC- Olkaria Volcanic Complex

OW- Olkaria Well

POOH- Pulling Out of Hole

ppm- parts per million

RIH- Running In Hole

RKB- Rotary Kelly Bottom

ROP- Rate of Penetration

TAS- Total Alkali versus Silica

TD- Total Depth

WOC- Wait on Cement

WDS- Wavelength Dispersive Spectrometer

Xps- Mole fraction of pistachite

XRD- X-ray diffraction

xix

Acknowledgements

My heart felt gratitude goes to my employer, Kenya Electricity Generating Company

Company Ltd. (KenGen) for nominating and providing me with the study leave to pursue

further studies at the University of Iceland. Sincere thanks goes to the Icelandic Government,

the United Nations University- Geothermal Training Programme (UNU-GTP) through its

able Director, Mr. Lúdvik S. Georgesson for funding this study. To the entire warm staff of

UNU-GTP, Ingimar Haraldsson, Thórhildur Ísberg, Málfríður Ómarsdóttir, Markús Wilde,

and Rósa Jónsdóttir, thank you for your technical support and for making me feel always at

home particularly during the harsh winter periods. Rósa you played a key role in delivering

the required research papers and books, even from outside the UNU-GTP library fairly fast.

I am deeply obliged to you and will always be.

I owe a large debt of gratitude to my supervisors; Guðmundur H. Guðfinnsson, Sæmundur

A. Halldórsson, Anette K. Mortensen and Björn S. Harðarson for the many insightful

discussions, constructive comments and guidance during my study period. Your countless

words of support utterly encouraged and motivated me. You helped me shape the line of my

thought with very many intellectual suggestions. It was a great pleasure and honour to work

with you and indeed I am very thankful. In the same token, I thank Dr. Hjalti Franzson and

Dr. Kristjan Saemundsson for the innumerable discussions we held together regarding the

various aspects of this study. Warm thanks are also extended to Atli Hjartarson for

preparation of the microprobe samples.

To my fellow UNU MSc. and Ph.D. students who form a long list, I am grateful to you all

for the times we spent encouraging each other and sharing common problems. In particular,

your ingenuity and hardwork generated the insights that taught me how the World of science

works. I recognize the unstinting support extended by my colleagues at KenGen in terms of

timely data delivery. Joyce Okoo, Xavier Musonye, Danson Warui, Michael Mwania, David

Wanjohi, Edwin Wafula and Lawrence Onyango, just to mention a few, you are indeed

awesome people and I am greatly indebted.

To my family, thank you for your encouragement, infinite patience during my long absence

and for allowing me a peaceful work time. Your love inspired me every day. Above all,

Glory be to the Almighty God. None of this work would have been possible without his

guidance and protection.

1

1 Introduction

1.1 General overview

Olkaria geothermal field is a high-enthalpy geothermal field (temperature > 200o C at 1000

m depth), associated with a region of Quaternary volcanism in the Kenyan Rift System

(KRS). The Olkaria geothermal field has been extensively explored and developed since the

1980s. Over the past 35 years, the geothermal field has been the study site for new and

innovative geological and geothermal research in Kenya. The geothermal system is located

in the Olkaria Volcanic Complex (OVC), a relatively young (~ 20 ka), salic volcanic

complex, characterised by the eruption of peralkaline rhyolites (Omenda, 1998; Macdonald

et al., 2008). Nine other similar

Quaternary volcanic centres

occur in the axial region of the

KRS and are potential

geothermal resources according

to surface exploration studies

(Simiyu, 2010). The OVC is

located in the central sector of

the KRS, ~ 125 km northwest of

Nairobi city and ~ 40 km

southwest of Lake Naivasha (0°

53’S; 36° 18’E) (Figure 1). The

extent of the OVC is estimated

to be ~240 km2 (Clarke et al.,

1990). It is bound in close

geographical proximity to the

north by Eburru, a

predominantly pantelleritic

volcanic complex, to the east

and south by Longonot and

Suswa volcanoes, respectively.

However, chemical features

indicate that all these four

adjacent volcanic complexes,

situated less than 50 km apart are

quite distinct and derived from

separate magmatic systems (e.g.

Macdonald and Scaillet, 2006).

Unlike Eburru, Longonot and

Suswa, which are complexes

with a distinct caldera, the OVC

lacks a proven caldera structure.

Figure 1: Map of the KRS showing the location of Olkaria

Volcanic Complex and other surrounding volcanic

centres (modified from Ofwona et al., 2006).

2

Various scientific studies have been conducted to establish the existence of the caldera

structure with little success. Some scholars (e.g. Simiyu and Keller, 2000) have postulated

that the group of coalesced rhyolitic domes in the eastern and southern segments of the

volcanic complex mark the rim of a caldera structure.

Other researchers (e.g. Bliss, 1979) have argued that the great thickness of comenditic lavas

(at least 1 km from borehole data) may well suggest the existence of a central comendite

volcano in Quaternary times. Such a view would be consistent with the voluminous

outpourings of comenditic lavas observed on the surface within the OVC. While this

interpretation seems possible, the situation is complicated by the lack of ignimbrite deposits

within and in the surrounding area of OVC.

Exploitation of geothermal resources at Olkaria began in the early 1980s and continues to

date with more than 200 wells having been drilled, with a depth range of between 950 and

3200 m. The Olkaria geothermal field has been divided into seven sub-fields with Olkaria

hill (Figure 2), a prominent geological feature about 340 m high and a basal diameter of 2

km being the reference point. Each of these sub-fields has been explored either partially or

completely. They include Olkaria east, Olkaria northeast, Olkaria northwest, Olkaria

southwest, Olkaria central, Olkaria southeast and Olkaria domes (Figure 2).

Olkaria east sub-field was the first part of the system to be developed in 1981. Exploitation

of the extensive geothermal resource at Olkaria is mainly for electricity production at four

conventional power plants; the Olkaria I, Olkaria I (AUs 4 & 5), Olkaria II, Olkaria III

(OrPower) and Olkaria IV power stations. In addition, there are several wellhead generators

installed and are operational within the geothermal field. As of the end of February, 2016,

Figure 2: The sub-fields of Olkaria geothermal field (modified from Otieno and

Kubai, 2013).

3

the aggregate installed generating capacity of the combined Olkaria stations is estimated at

~ 654 MWe. Resource consent for a 140 MWe power plant development at Olkaria V was

granted by a Board of Inquiry process in May, 2015, with inauguration of construction work

slated to begin in the course of this year. Although other opportunities for direct use are

being explored, the activity is currently largely restricted to greenhouse heating (i.e. Oserian

Development Company) and balneology (Olkaria Spa). A summary of the power plants,

including the year of commissioning and the effective installed capacity, is listed in Table 1.

Table 1: Utilization of geothermal energy for electric power generation in Olkaria.

Power Plant

Name

Year

Commissioned

No. of

Units

Status Total Installed

Capacity

Total Running

Capacity

MWe Mwe

Olkaria I 1981 (15 MWe)

1982 (15 MWe)

1985 (15 MWe)

3 Operating 45 45

Olkaria II 2003 (70 MWe)

2010 (35 MWe)

3 Operating 105 105

Olkaria I (AUs

4 & 5)

2015 2 Operating 150 140

Olkaria III

(OrPower)

2000 (53 MWe)

2008 (39 MWe)

2014 (17 MWe)

2016 (30 MWe)

- Operating 150 140

Olkaria IV 2014 2 Operating 140 140

Olkaria

wellheads

2012- 2015 - Operating ~ 64 ~ 64

1.2 Overview of well OW-922

This study largely focuses on well OW-922, in addition to wells OW-905A, 910 and 917.

Well OW-922 is a step-out vertical well drilled between April and September, 2014 at

surface coordinates X= 208675; Y= 9899449 and Z= 2126 masl. It is located approximately

2.5 km directly east of the proposed Olkaria ring structure (Figure 3). The purpose of drilling

this well was to appraise the resource potential of the mentioned area so as to expand

production drilling to the plain outside the ring structure. This has been necessitated by the

renewed focus by the Kenyan government on development and utilisation of the geothermal

resources for electricity generation. The government has identified additional supply of

renewable energy as a fundamental foundation of achieving its Vision 2030 blue print. The

latter seeks to transform the country into a newly industrialised, middle-income economy.

As a result, Kenya Electricity Generating Company Ltd. (KenGen), the leading power

producer in the country, has embarked on efforts to escalate development of geothermal

resources in sync with its geothermal expansion programme. On the other hand, wells OW-

905A, 910 and 917 were drilled within and along the proposed Olkaria caldera structure

between November, 2007 and December, 2012. These wells have been adopted for this study

based on various aspects; specifically owing to the readily available data on borehole

geology and sub-surface petrochemistry from a previous study (Musonye, 2015). Until

4

recently, there were no prior sub-surface petrochemical data (major and trace elements) since

past petrochemical studies were confined only to surface rocks (e.g. Omenda, 1997;

Macdonald et al., 2008; Marshall et al., 2009).

1.3 Goals of the study

The specific aims of this study are three fold;

1. To study the borehole geology of well OW-922 by analysis of drill cuttings in order to

identify the stratigraphy and the sub-surface geological formations. The borehole

geology study will clarify the geothermal conditions in the system east of the Domes

and lead to comprehensive understanding of the hydrothermal mineralisation,

temperature conditions based on alteration minerals and temperature loggings, major

sources of permeability and water-rock interaction processes.

2. To investigate the chemical characteristics of the rocks in well OW-922 and attempt to

decipher the processes of their petrogenesis. These will be compared with petrochemical

data from wells OW-905A, 910, 917 and surface samples from Olkaria and neighboring

minor eruption centres to establish any petrogenetic connection.

Figure 3: A section of the Olkaria Domes field showing location of the step-out

well OW-922 and other study wells OW-910, 917 and 905A. Also shown are other

drilled wells. The black triangles delineate the outline of the proposed ring

structure.

5

3. To assess the chemical composition and variation in compositional range of epidote and

chlorite, as well as the variables likely influencing the compositional changes.

1.4 Thesis layout

Chapter 1: In this chapter, the project area is introduced, including the sub-divisions, total

installed capacity and location of the study wells. The objectives of this study and the outline

of the thesis are presented.

Chapter 2: Provides an overview of the regional geology of the East African Rift System.

An elaborate description of both surface and sub-surface geology, structural and tectonic

setting, hydrogeological setting and reservoir characteristics of the study area. Also

presented is the drilling history of well OW-922.

Chapter 3: Provides an overview of the analytical methods used to achieve the project

objectives. Analytical techniques spanning from binocular microscope to electron

microprobe analysis are described.

Chapter 4: Presents synthesis of borehole geology, whole-rock chemistry and electron

microprobe analytical results. In borehole geology, the lithologic units intercepted by the

drill hole are discussed. Also discussed are the distribution of hydrothermal mineralisation,

feed zones and a correlation between inferred alteration and measured formation

temperatures. In whole-rock analysis, a petrochemical classification of the sub-surface rocks

in the study well and the compositions of major and trace elements are explained in detail.

Chlorite and epidote chemical compositions derived by electron microprobe (EMP) analysis

are detailed.

Chapter 5: Provides an in-depth discussion based on the outcome of the analytical results.

Chapter 6: This chapter incorporates the conclusions based on the outcome of the findings

envisaged in all the analyses and future frontiers recommended.

7

2 Outline of Geology

2.1 Regional geology

The East African Rift System (EARS) is considered an archetypal continental rift in the

initial stage of continental breakup (Achauer and Masson, 2002) that is still in close

proximity to the underlying thermal anomaly (mantle plumes). The evolution of the EARS

is strongly believed to be structurally controlled (Smith and Mosley, 1993) by the

exploitation of weak

collisional zones at the contact

between the Proterozoic

orogenic belts and the Archean

Tanzanian Craton by the rift

faults (Shackleton, 1996).

South of Turkana (Figure 1),

the rift is divided into two

distinct branches that encircle

the robust Tanzanian Craton:

an older, volcanically active

eastern branch and a younger,

much less volcanic active

western branch (Ring, 2014)

(Figure 4). The eastern branch

extends from the Red Sea in

the Afar region, straddles

through the Main Ethiopian

Rift (MER), the Kenyan

(Gregory) Rift and the basins

of the Tanzania divergence

before terminating further

south in Beira, Mozambique, a

total distance of more than

4000 km. By contrast, the

western branch extends over a

distance of more than 2100 km

from Lake Albert (Mobutu) in

the north to Lake Malawi

(Nyasa) in the south

(Chorowicz, 2005).

The zone is typically trough-like, ~ 40 to 65 km wide, and traverses two broad, elongated

domal uplifts in Ethiopia and Kenya. The elevation of the rift floor is highest in the central

part of the domes, 2000 m in Kenya and 1700 m in Ethiopia, and decreases progressively to

the fringes. The onset of topographic uplift in the EARS is poorly dated. However, it

certainly preceded graben development (Baker et al., 1972). The evolution of EARS dates

Figure 4: DEM of the EARS. Black lines represent main

rift faults, whereas, white surfaces are lakes (modified

from Chorowicz, 2005).

8

back 30-45 Ma (Ring, 2014). Rifting is generally assumed to have developed as a result of

the upwelling of a voluminous mantle plume (the African Superswell), which continuously

impinged upon the base of the stretched continental lithosphere (Ebinger and Sleep, 1998).

The driving force for the process has been ascribed to a hot, relatively fertile asthenospheric

plume or diapir emanating from the top of the African Superswell in the upper few hundred

kilometres of the mantle (Ebinger and Sleep, 1998). To date, considerable uncertainty still

exists about the number of plumes influencing magmatism and tectonics beneath the EARS.

Specifically, the depth extent and continuity of the hot asthenospheric material remains a

tantalizing question which has evoked much speculation and debate (Chorowicz, 2005).

Based on plate analysis, the rift marks a line of very slow crustal spreading, extension being

more active in the Red Sea – Gulf of Aden area, approximated at ~ 2 cm/year. In the Main

Ethiopian rift it is estimated at ~ 1

mm/year, and further less than 1

mm/year in the Kenyan Rift and

southwards (Ring, 2014).

Propagation of the rift began as a

chain of marginally warped

depressions, which were

accentuated as domal uplift

proceeded, until, in mid-Miocene

to early Pliocene times, faulting

produced asymmetrical grabens.

The final uplift phase in the early

Pleistocene was accompanied by

major graben faulting, and

subsequent faulting has intensely

fractured the floor of the rift along

an axial zone marked by caldera

volcanoes. Seismic velocity

information from refraction

measurements and tomography

experiments, reinforced by

gravity studies (Ebinger, 2005),

indicate that the eastern rift lies

along a zone of progressive

crustal thinning with local crustal

disruption.

According to Baker et al. (1972) the uplift of the Ethiopian and Kenyan domes (Figure 5)

has been synchronous in three major pulses of late Eocene (44-38 Ma), mid-Miocene (16-

11 Ma), and Plio-Pleistocene (5-0 Ma) age. Volcanism of intermediate and salic type

displays some relation to uplift in time and space and to the onset of graben faulting.

However, major flood basalt extrusions in the early Tertiary in Ethiopia were related to

massive crustal warping along the future rift margins (Williams, 1970). The volcanism

associated with EARS is overwhelmingly alkaline, and at some volcanoes a strongly alkaline

fractionation series is distinguishable from a more mildly alkaline series (Williams, 1970).

The flood phonolites, trachytes, rhyolites, and ignimbrites of Kenya, and the pantelleritic

Figure 5: Map of the EARS showing elevations higher

than 1200 m (black areas), evidencing the main

Ethiopian and Kenyan Plateaux (dashed lines). The

black areas mark the Cenozoic rocks of the Rift

(Macdonald et al., 2001).

9

ignimbrites of Ethiopia, could have resulted from anatexis of a mantle-derived accreted layer

at the base of the crust (Baker and Wohlenberg, 1971).

The evolution of the KRS, a complex graben ~ 40-65 km wide bounded by major rift faults

arranged en echelon, dates back to the East African orogeny. The rift bisects the Kenya

domal uplift, which itself is superimposed on the eastern margin of the East African plateau

(Baker et al., 1972). The KRS is also located close to the boundary of the Tanzanian Craton

and the Pan-African Mozambique shear belt. Volcanism associated with the KRS started

during the Miocene epoch (~ 24 Ma), resulting in widespread basaltic eruptions in the

Turkana topographic depression, a failed Mesozoic rift system (Nyblade and Robinson,

1994). The Miocene basalts were subsequently faulted and followed by the relatively

extensive, reliably dated flood phonolites. The latter are of chemically homogeneous

composition and were erupted from the crest of the Kenya dome in the late Miocene and

early Pliocene (Baker et al., 1972). The total volume of eruptive rocks is estimated to be >

220,000 km3, with a thickness of up to 900 m, as revealed by faulting at the rift margins. In

an important contribution, Ebinger and Sleep (1998) suggest the flood phonolites to be a

product of partial melting of the lowermost part of the crust.

Volcanism off the rift axis was concurrent with the rifting process and is responsible for the

formation of the vast, uplifted, off-rift volcanic centres, namely Mt. Kenya, Chyulu and Huri

hills, located on the eastern flanks (Figure 1). It has been observed that these three sites of

domal uplift are not rifted, and, support a model whereby doming predated rifting (Wilson,

1989). Further block-faulting of the Miocene volcanics produced noticeably massive and

extensive, more evolved eruptions of trachytic ignimbrites in the central area, provisionally

assigned to lower Pliocene (Ring, 2014). The trachytic ignimbrites formed the Mau and

Kinangop tuffs. A third faulting episode, which followed the ignimbrite eruptions, resulted

in the formation of the graben structure, as it is known today (Baker et al., 1972). In the

developing graben, successive fissure eruptions produced approximately 900 m thick layers

of trachytes, basalts, basaltic trachyandesites and trachyandesites. The plateau rocks that

filled the developing graben were subsequently block-faulted to create high-angle normal

faults, which are common in the axial graben of the rift floor. The fractures apparently served

as conduits for a series of Quaternary volcanoes of basaltic to silicic composition (Ebinger

and Sleep, 1998).

OVC is a classical example of such a Quaternary volcano of the KRS. Characteristic

volcanism within the OVC and other centres exposed on the periphery of the complex (e.g.,

Oserian, Kibikoni, Olenguruoni, Broad Acres and the south-westerly Arcuate Domes)

(Figure 8) are dominated in outcrops by peralkaline rhyolite domes and lava fields. The range

of the stratigraphic units observed in each centre is variable. Similarly, each eruption centre

corresponds to a different age as discussed by Marshall et al. (2009). Two main magmatic

differentiation processes, namely crustal anatexis and crystal fractionation, have been

proposed by Davies and Macdonald (1987) to explain the generation of felsic rocks at OVC

and along the Gregory Rift at large.

2.2 Geology of Olkaria Volcanic Complex

Knowledge of the sequence, nature and architecture of the stratigraphy is fundamental for

understanding both the surface and sub-surface geology of the OVC. The current knowledge

10

on the geology of OVC and its surroundings presented hereafter, derives from the detailed

pioneering work of many scholars (e.g. Clarke et. al., 1990; Marshall et al., 2009; Omenda,

1998). Their work has provided the foundation upon which more recent work has continued

to build and has been refined through subsequent field development. The OVC constitutes

part of the Central Kenya Peralkaline Province (CKPP) (Macdonald and Scaillet, 2006),

which coincides with the Kenya Dome; a ~ 300-400 km wide topographic culmination in

the course of the rift, with the apical region postulated to be lying around the Lake Nakuru

area (Figure 1) (Macdonald and Scaillet, 2006). Nearly 80 volcanic centres composed of

lavas and/or pyroclastic rocks formed by eruptions through fault zones or occuring as steep-

sided domes are found within the volcanic centre, hence the term complex. Clarke and co-

workers (1990) have notably placed the evolution of the OVC to be in the period between

22-20 ka BP.

2.2.1 Surface geology

The surface lithologic units in OVC are dominated by comenditic lavas, pumice fall and

pyroclastics (Figure 6). A large fraction of the pumice fall and pyroclastic deposits is

hypothesized to have originated from Longonot and Suswa volcanoes, lying immediately 20

km east and 40 km south of OVC, respectively. However, there is need for geochemical

analysis of the deposits from the three volcanic centres to establish this assertion. Even

though it has not been possible to enumerate systematically the contribution of pyroclastic

deposits from each of the three centres, the pyroclastic activity at Longonot is presumed to

post-date volcanism at Olkaria (Omenda, 1998). Occurence of the comendites is restricted

to OVC and it is the only area in the whole of the KRS with these rocks on the surface.

Figure 6: Geological map of the Olkaria Volcanic Complex and the surrounding

area (modified from Clarke et al., 1990).

11

Clarke et al. (1990), listed six stages in the evolution of the complex based on their geological

work. A tabulation of the sequence of volcanic groups is provided in Figure 7 and their

distribution summarised in ascending order as discussed below. On the other hand, Olkaria

eruptive centres and their manifestation as distinct topographical units are shown in Figure

8.

Stage 1: The earliest stage, of uncertain age, resulted in a pile of dominantly trachytic lava

and pumice. These are termed Olkaria Trachyte Formation (Ot) and Maiella Pumice

Formation (Mp). The former is exposed mostly along ridges and gullies in the southwest

portion of the complex, whereas the latter is thought to have been erupted from the extensive

vents in the western segment of the complex.

Stage 2: This stage, also of uncertain age is believed to mark the collapse of the proposed

caldera structure, leaving behind a depression 11 × 7.5 km across. The suggested caldera

formation was preceded by an eruption of a large volume of densely welded pantellerites of

the Ol-Njorowa Pantellerite Formation (O1). The latter are mainly exposed in the deepest

sections of Ol-Njorowa gorge (Figures 6 and 7). Based on field observations, Clarke et al.

(1990) relied on a group of coalesced rhyolitic domes (Figure 8), forming distinct

topographic features in the eastern and southern parts of the complex to infer the ring

structure. However, the caldera structure is still poorly constrained because the curved

lineament of rhyolitic domes is adjudged to demarcate only 30% of a caldera rim.

Stage 3: This stage was characterised by early post-caldera activity and is represented by the

eruption of pyroclastic rocks (Op2) and peralkaline rhyolitic lavas (O2) of the Lower

Comendite Member of the Olkaria Comendite Formation. The formation has been dated by

Clarke et al. (1990) at > 9150±110 BP.

Stage 4: Involved ring dome building (O3) and the eruption of thick surge deposits (Op3)

representing the Middle Comendite Member. 14C dating places the age for this member at

between 9150±110 BP and 3280±120 BP (Heumann and Davies, 2002).

12

Stage 5: General resurgence of the caldera floor lead to the formation of short, thick

comendite flows of the Upper Comendite Member (O4). This has been dated at >3280 BP

(Clarke et al., 1990).

Stage 6: This stage was centred on the N-S Ololbutot fissure system (Figure 6). The most

notable expression of this stage is very thick comenditic flows of the Ololbutot Comendite

Member (O5). The youngest of these flows is the Ololbutot lava, dated at 180±50 BP by 14C

(Macdonald and Scaillet, 2006).

Apart from the stages of evolution of the OVC, Clarke et al. (1990) also recognized a series

of eruptive mafic rocks, occuring both to the north and south of the Olkaria Domes field.

The Lolonito Basalt Formation (Ba1) is found to the south (Figure 7), forming a string of

small outcrops and largely comprised of pyroclastic cones and flows. The age of this

formation has been estimated at < 0.45 Ma and it typifies an early phase of the younger Akira

Basalt Formation (Ba2). The Tandamara Trachyte Formation (Tt), lying in the immediate

vicinity of Akira Basalt Formation to the southeast, is postulated to be approximately coeval

with the Akira Basalt Formation (Clarke et al., 1990). To the north of the domes field, there

is widespread mafic volcanism associated with rhyolite emplacement on the expansive

Ndabibi plains (Figure 7). The latter is a low-lying area between the OVC and the Eburru

Volcanic Centre. These have given rise to the Ndabibi Basalt Formation (Bn), comprising

primarily faulted lavas and pyroclastic cones of pre-caldera age. The age range for the Bn is

Figure 7: Surface formations and stratigraphic column of Olkaria Volcanic

Complex. Also shown are a series of eruptive peripheral mafic and trachyte

formations (Macdonald et al., 2008; Marshall et al., 2009).

13

uncertain, a fact attributed to some lavas having well-preserved topography and lacking

vegetation, possibly suggesting that they are young. Nonetheless, some of the Bn flows are

intersected by minor faults and therefore considered older.

2.2.2 Sub-surface geology

Drill cuttings from several geothermal wells have contributed valuable information to the

understanding of the sub-surface strata of the Olkaria geothermal field. In particular, drilling

to depths > 2000 m has provided new insights into the deep sub-surface stratigraphy of the

Olkaria geothermal field by enabling detailed geological analysis of drill cuttings. The litho-

stratigraphic units of the Olkaria geothermal area, as revealed by data from geothermal wells

Figure 8: Geographical extents and stratigraphic units of the Olkaria

eruptive centres. The Ol-Njorowa Pantellerite Formation (O1) is not

shown (Marshall et al., 2009).

14

and regional geology, have been classified into six distinct groups. These groups are based

on age and lithology as discussed in the following section (from the youngest to the oldest).

The Upper Olkaria Volcanics are stratigraphically the youngest unit within the complex

(Omenda, 1998), dated at < 0.95 Ma. Pyroclastics and comendites are prevalent within this

series with minor interbeds of trachyte and basalts (Figure 9). Rocks between the surface

and ~ 1500 m.a.s.l are part of this formation, the youngest being the aforementioned

Ololbutot lava flow dated at 180±50 yrs ago (Macdonald and Scaillet, 2006). Moreover, the

comendites forming a ring of domes east of the complex and those exposed along the Ol-

Njorowa gorge belong to this series. Nonetheless, the comenditic domes have been masked

by the comparatively thick pyroclastic flows deposited in the Domes field and surrounding

area. Some of the deposits are believed to have originated from the adjacent Longonot and

Suswa volcanoes.

The Upper Olkaria

Volcanics formation is

underlain by Olkaria basalts.

Basaltic lavas are

predominant, though

alternating with thin

horizons of tuffs, minor

trachytes and sporadic

rhyolites. The formation is

extensive in the East,

Northeast, Southeast and

Domes fields but absent in

the West field. The

formation has been

intersected by wells at

between 1900 and 1500

m.a.s.l (e.g. Okoo, 2013)

and forms the cap rock for

the Olkaria geothermal

system. Besides, the

formation has been used as a marker horizon within the geothermal field, especially in the

Domes field. The Olkaria basalts are preceded by Plateau Trachytes of Pleistocene age. The

trachytes form part of the Kenya rift floor fissure flows which are exposed to the south and

north of OVC. The formation is characterized by trachyte as the principal rock type with

localized occurence of tuffs, rhyolites and basalts. Plateau trachytes is the reservoir rock for

the East, Northeast, Southeast and Domes fields. Its thickness has been estimated to be

greater than 1.5 km based on borehole data from these fields (e.g., Mwangi, 2012, Okoo,

2013).

The Mau tuffs, dated between 3.4-4.5 Ma, are the oldest rocks encountered within the

complex and are of unknown thickness. The formation has been penetrated by drill holes in

the west fields (Northwest and Southwest), where it forms the reservoir rock. However, it

has not yet been intersected by wells in the other Olkaria fields. The lithotypes constituting

the Mau tuffs are composed almost exclusively of tuffs with minor interbeds of rhyolites,

basalts and trachytes. Pre-Mau volcanics, dated at > 4.5 Ma constitute the fifth series and

Figure 9: Sub-surface lithotypes of the Olkaria Volcanic

Complex (Modified from Omenda, 1998).

15

overlie the basement rock. The lithotypes in this series consist notably of phonolites, basalts,

trachytes and tuffs. Underlying the Pre-Mau volcanics is a laterally extensive basement

system comprising Proterozoic metamorphic amphibolite grade gneisses and schists,

accompanied by marble and quartzites of the Pan-Africa basement system. This unit has

been dated at > 590 Ma. The depth to the basement has been interpreted by Simiyu and

Keller (2000) to be about 5-6 km in the central Kenya Rift. The apparent absence of these

two units, both on the surface and within the sub-surface geological units of the Olkaria

geothermal system, has been interpreted by Shackleton (1996) to be a consequence of a

major graben trending N-S towards the southern flanks of the rift.

2.3 Structural geology and tectonics of OVC

Several researchers have studied and documented the distribution of geological structures in

the OVC and surrounding areas since 1971 (e.g. Baker and Wohlenberg, 1971; Omenda,

1998). As mentioned earlier, the OVC is situated in the cental sector of the elliptically shaped

Kenya Rift, a region marked by a sharp re-orientation in the rift’s trend from a NNW

alignment, to the north of the volcanic complex, to a more or less SSW direction to the south

of the complex (Figure 1). Some dispute still exists regarding the explanation for the change

in the rift’s trend, extending for nearly 120 km before resuming a SSW course. For instance,

Smith and Mosley (1993) have proposed this bend as being an intersection with a large NW-

striking basement structure, the ‘Aswa lineament’ (Figure 1). The theory of the involvement

of the basement structure has, however, been rejected by, inter alios, Bosworth (1987) who

ascribed the shift in the rift orientation to represent an accommodation zone. The latter is

taken to separate two-half grabens, the southern Magadi-Natron sub-basin and the central

Nakuru-Naivasha sub-basin.

An intimate knowledge of the structural controls and their topological relationships in the

OVC is instrumental in determining the reservoir fluid flow at depth, and consequently, in

well siting. The determination of breccia zones associated with faults that could act as

channels for geothermal fluids is of key importance. According to Steiner (1977), some

faults have proved to be productive and serve as efficient conduits for the ascent of

geothermal fluids, while others are known to be totally unproductive. Geological structures,

especially faults and fractures, in the Olkaria geothermal system obviously play a significant

role by serving as pathways for the geothermal fluids. Hence, they are the primary means

for efficient heat transfer from deep to shallower levels of the system. Additionally, the

structures have apparently displaced some geological units within the geothermal system, as

witnessed, for example, by the presence of the Mau tuffs in Olkaria West field but lacking

in wells in the East field. The structures in OVC are divided into rift faults, ring structure,

Ol-Njorowa gorge, and dyke swarms (Figure 10).

16

The structural domains of the OVC are dominated by NE-SW and NW-SE striking faults,

and subordinate N-S, NNE-SSW and ENE-WSW structural trending patterns (Figure 10).

Other faults of variable orientation, though limited, also occur in the area and are not

included in the structural map. Faults are more conspicuous in the East, Northeast and West

Olkaria fields. However, their presence on the surface in the Domes area is severely limited

by the substantial blanketing of the surface geology by thick, younger pyroclastic deposits

(Otieno et al., 2014). The oldest faults of the complex are the NW-SE (e.g. Gorge farm fault)

and ENE-WSW (e.g. Olkaria fault zone) trending faults (Omenda, 1998), and are associated

with the rift formation (~ 30-45 Ma). The most recent fault systems, i.e. the N-S, NE-SW

and NNE-SSW trending ones, are considered to be of a different structural regime and are

thus interpreted to be associated with later tectonic activities, and perhaps, the proposed

caldera collapse. The most obvious of these structures are the Olkaria fracture and Ololbutot

fissure (Figure 10). The trace of the latter is revealed by extensive hydrothermal

manifestations (mainly fumarolic activity), characterized by deposits of kaolinite and silica.

This fault has obviously an impact on the vertical permeability witnessed along the N-S trend

(Mungania, 1999).

Surface geothermal activity in Olkaria coincides with fault arrays. Prominent manifestations,

notably, hot springs, fumaroles and altered grounds, display a strong relationship with the

tectonic lineaments in the area. The Ol-Njorowa gorge, an erosional channel extending for

~16 km in total length, has been dated at about 9 ka BP (Clarke et. al., 1990). It is aligned

roughly in a NE-SW direction and marks the boundary between the East and the Domes

Figure 10: Structural map of the Olkaria Volcanic Complex displaying the orientation

of various fault patterns (modified from Omenda, 1998).

17

fields (Figure 10). Studies by Clack and co-workers (1990) suggest the gorge to be a product

of faulting and was later deepened by catastrophic southward overflow of Lake Naivasha

when the lake level was much higher than at present. Numerous felsic dykes trending NNE-

SSW and volcanic plugs (necks) are widely distributed along the gorge, strongly suggesting

that faulting played a key role in its initiation (Otieno et al., 2014). Further evidence is

provided by the active, boiling fumaroles, possibly attesting to reactivation of some of the

faults along the gorge.

The arcuate alignment of rhyolitic domes in the eastern and south-eastern parts of Olkaria

has been interpreted by Omenda (1998) to indicate the presence of a buried caldera. In his

hypothesis, he proposed the arcuate alignment of rhyolitic domes and ridges to be products

of a resurgence. Perhaps a more plausible explanation is that the ring structure resulted from

magmatic stresses in the Olkaria magma chamber (Omenda, 1998). Nonetheless, there is no

overwhelming evidence for any of the many possible explanations. As highlighted earlier,

the exposure of geologic surface structures is limited in the Domes area and its immediate

vicinity due to the cover of the surface by pyroclastic deposits. Most of the structures in this

area are sub-surface and have been deduced through surface mapping along the deep gullies

and well drilling operations. Evidence for buried faults, has been found through stratigraphic

correlation of wells (e.g. Lagat, 2004) owing to the presence of excellent marker beds (Figure

9). Cave-ins and total loss of circulation returns during drilling operations, for instance in

well OW-915B (KenGen, 2012), have also served as a clear indicator for the existence of

sub-surface fault structures.

2.4 Hydrogeological setting

The hydrogeology of OVC, and to a greater extent the surrounding volcanic centres (i.e.

Eburru, Longonot and Suswa) within the central sector of the KRS, is controlled primarily

by the rift faults. Tectonic movements of the rift undoubtedly played a pivotal role both on

a large-scale, by forming the regional faults, and on a small-scale, by creating the local

fracture systems of enhanced permeability. These faults and fracture systems act as

permeable pathways by channelling lateral groundwater flow from the recharge areas into

the low-lying geothermal systems along the rift axis. Ordinarily, faults have the capacity to

propagate fluid flow by providing zones of high permeability. Likewise, faults may

simultaneously offset zones of relatively high permeability (Allen et al., 1989), in which

case they cause compartmentalization of lateral flow.

Recharge into the geothermal field at Olkaria is exclusively of meteoric origin (Ojiambo and

Lyons, 1993), derived from high altitude precipitation at the rift escarpments (Mau

escarpment to the west and Kinangop plateau to the east) (Figure 11). The two highlands

have a maximum elevation of 3080 m and 2740 m, respectively, and experience an annual

rainfall of between 1200-1500 mm. On the other hand, Olkaria geothermal field and its

environs receive an estimated average annual rainfall of about 826 mm (Allen et al., 1989).

The structures controlling deep recharge within the system are the NW-SE and NNW-SSE

trending faults (Figure 10). As previously reported (Ambusso and Ouma, 1991), the

rejuvenated pre-Pleistocene, ENE-WSW trending Olkaria fault zone (Figure 10) is a major

structural divide transecting the Northeast and West fields. Therefore, it is regarded as an

important structure forming a hydrologic divide within the geothermal system. This, for

instance, has been demonstrated by the presence of a liquid-dominated reservoir without a

18

steam cap in the north and a liquid dominated two-phase reservoir overlain by a steam cap

in the south of the fault (Ambusso and Ouma, 1991). Some water from the recharge areas

also enters the Olkaria geothermal system through the late-Pleistocene, relatively young,

tectonically reactivated N-S faults and fractures (e.g. Ololbutot fault and Olkaria fracture

zone, respectively). These structures control the axial groundwater movement through the

geothermal system, though have a shallower influence in comparison to the major rift faults

that offer deep recharge (Omenda, 1998).

Other features which have been noted to enhance fluid flow within the Olkaria system, and,

to a larger extent in the Domes area, are lithological interfaces and, jointed and fractured

lavas. Similarly, stable isotope studies have been employed (e.g. Nkapiani, 2014) to

determine the origin of thermal waters and the reservoir recharge areas of the Olkaria

geothermal system. Based on hydrogen isotope studies, the reservoir fluid isotope

composition of the thermal waters indicate two potential recharge zones for the system; the

eastern rift flank (δD = ~ -24‰) and the western rift flank (δD = ~ -30‰). Results of

hydrogen isotope determinations suggest the West, East and Northeast fields to be recharged

by waters from the western escarpment. The Domes field obtains recharge largely from the

eastern rift wall (Nkapiani, 2014). Whether or not sub-surface drainage from Lake Naivasha

Figure 11: Piezometric map showing reservoir recharge areas

for Olkaria and other surrounding geothermal fields along the

rift axis (modified from Allen et al., 1990).

19

contributes to recharge of the Olkaria geothermal system has intrigued many researchers and

is still a subject of debate (e.g. Allen et al., 1989).

2.5 Reservoir properties of Olkaria field

In this section, attention is focused principally on the most important features of the reservoir

system of the Olkaria geothermal field. A comprehensive review of all the aspects of

reservoir characteristics of the Olkaria geothermal field is beyond the scope of a single sub-

section. The Olkaria geothermal reservoir has developed as a consequence of the OVC.

Presently, the geothermal system supports ~ 654 MWe (installed capacity) through power

stations operated by KenGen and Ormat Technologies (OrPower) (Table 1). It is a two-phase

liquid dominated system overlain by a comparatively thin steam-dominated zone, ranging

from 100 to about 200 m in thickness at 240oC. The hydrogeological features of the field are

distinguished by four up-flow zones in the general area. Extensive fluid circulation in the

field is controlled by major faults and fracture zones as discussed earlier. The ultimate heat

source of the system has been attributed to a deep-seated magma chamber, which peaks in

several locations in the form of three main intrusions, possibly partially molten. Based on

seismic studies by Simiyu and Keller (2000), the intrusions extend to shallower levels,

approximately 6-8 km, and are suggested to lie beneath Olkaria Hill, Gorge Farm volcanic

centre and in the Domes area. Temperature and pressure models supports this observation

and, have identified four main up-flow zones related to these heat sources (Figure 12)

(Mannvit consortium, 2011).

Figure 12: A vertical schematic sketch of the conceptual model of

Olkaria geothermal system. The suggested heat sources, up-flow zones

and temperature cross-section are shown. The green arrow is directed

north (Mannvit consortium, 2011).

20

The first up-flow zone feeds the Olkaria west fields and is linked to the Olkaria Hill (Figure

2) heat source body. The second up-flow zone, feeding the Northeast and East fields as well

as the northwest corner of the Domes field, is connected to the heat source beneath the Gorge

Farm volcanic centre (Figure 8). The final up-flow zone, associated with the ring structure,

is visible in the southeast corner of the Domes field and is linked to the heat source suggested

beneath the area. Reservoir chemistry data support the existence of the up-flow zones as

defined by relatively high Cl- concentrations and high Na/K temperatures within these zones

(Mannvit consortium, 2011). Above the steam zone, basaltic rocks, and to a lesser extent

trachytes, form an impermeable caprock, subsequently marking the top of the reservoir. This

proposition has been supported by reservoir modeling (Ambusso and Ouma, 1991), which

reveals sharp temperature increases below this formation. The depth to the caprock,

however, varies from one sub-field to another, but in many instances lies between 1900 and

1300 m.a.s.l, as revealed by several stratigraphic correlations (e.g. Okoo, 2013; Mwania et

al., 2013). The bulk of the reservoir rocks, save for in the West fields, are trachytes of

Pleistocene age with well developed fractures. The vertical extent of the reservoir is quite

uncertain and varies from one sub-field to another. Ambusso and Ouma (1991) have

suggested it to be of the order of several hundred meters. However, the presence of extensive

intrusive bodies (e.g. syenite, granite, etc) at greater depths (e.g. below 2500 m) probably

indicates proximity to the bottom of the reservoir.

2.6 Summary of drilling history

Well OW-922 is a vertical well drilled to a total depth of 2990 m from the cellar top (CT),

to the east at approximately 2.5 km from the proposed ring structure. The well was spudded

on 16th of April and completed on 19th of September, 2014, after a period of 156 days. This

translates to 101 more days than planned, since the well had initially been expected to be

completed on 12th June, 2014. The drilling stages comprised the surface, intermediate and

production hole as well as a production section. Delays in the drilling were occasioned by

various challenges as outlined in the description of different phases. Casing depths were

measured from the flange. A summary of the drilling phases and casing design of well OW-

922 is illustrated in Table 2 and the drilling progress is shown in Figure 13.

Table 2: Drilling phases and casing design for well OW-922.

Drill Rig

Stage

Hole

diameter (")

Depth (m) Casing

From To Casing

diameter (")

Type

GWDC-116

Phase 1

Phase 2

Phase 3

Phase 4

26

171/2

121/4

81/2

0

58

300

1200

58

300

1200

2990

20

133/8

95/8

7

Surface

Anchor

Production

Slotted

liners

21

2.6.1 Phase 1: Drilling of the Surface hole

Surface hole was drilled with a 26" roller cone milled tooth bit using high-viscosity bentonite

based-mud to a depth of 58 m rotary kelly bottom (RKB). Full circulation returns occured at

the surface throughout the drilling of this phase. The well was circulated to the bottom and

no obstructions were detected. It was cased using a 20" outside diameter (OD) surface

casing. Casing shoe was set at 57.1 m RKB and the casing column successfully cemented.

This phase lasted 4 days.

2.6.2 Phase 2: Drilling of the Intermediate hole

Drilling of phase 2 started on 20th April, 2014 using a 171/2" roller cone bit with aerated

water and foam as the drilling fluid. Like during the drilling of the surface hole, full

circulation returns were received on the surface down to the section total depth (TD) of 300

m RKB. The hole was circulated before the 133/8" anchor casing was run in hole (RIH) and

casing shoe set at 299.19 m RKB. The casing column was cemented in place in three stages;

the primary cementing and two backfills. Drilling of this phase was completed in 6 days.

Figure 13: Drilling progress curve for well OW-922 (KenGen,

2014).

22

2.6.3 Phase 3: Drilling of Production hole

Drilling of this phase began on 26th April, 2014 with the use of a 121/4" roller cone bit.

Drilling progressed reasonably well with aerated water and foam being used with good

circulation returns realized at the surface. However, several unusual events occurred. On 29th

April, the drill string got stuck at a depth of 603 m RKB. Several attempts to free the string,

including pumping a mixture of PIPE-LAX and diesel, proved successful after 2 days. A

decision was then made to pull out of hole (POOH) and inspect the drill strings, before RIH

to continue drilling. Further attempts to proceed with drilling proved futile due to a second

obstruction encountered at 520 m RKB. At this point in time, the well condition had become

complicated. The crew embarked on reaming the wellbore from surface to bottom with

aerated water and foam, receiving no circulation returns at the surface. This was done in an

attempt to clear any obstructions along the well course.

While reaming almost reaching the bottom (603 m RKB), the drill string got stuck once

again. Efforts to free the string were immediately initiated. This was on 1st May, 2014. To

overcome this impediment, a decision was made to carry out a plug job. A total of 24 plug

jobs were conducted around this section. The decision proved successful and the section was

finally by-passed. Normal drilling commenced again on 29th May, with the use of mud and

progressed smoothly to TD of 1200 m RKB. Circulation of the wellbore and a wiper trip

were undertaken without any considerable challenges. RIH of the 95/8" OD production

casings followed and the casing bottom was set at 1197.86 m RKB. The casing column was

then cemented in place in three stages; the primary cementing and two backfills. Drilling of

this phase lasted 55 days.

2.6.4 Phase 4: Drilling of Production section (Main hole)

Drilling of the production section resumed on 20th June, 2014 with the use of a 81/2" roller

cone bit, with aerated water and foam as the drilling fluid, resulting in intermittent circulation

returns. Drilling progressed without any major hindrance before an obstruction was

encountered at 1359 m RKB. This was successfully overcome after 1 day of reaming, and

drilling continued smoothly with excellent circulation returns. On 8th July, (84th day) a tight

hole was encountered at 1280 m RKB while RIH with a new drill bit. A decision was made

to ream the wellbore with aerated water and foam before resumption of drilling, which

proved fruitful. The final depth of 3000 m RKB was achieved on 13th July, 2014 (89th day).

The crew embarked on circulation to unload the wellbore before RIH hole the 7" perforated

liner. However, an obstruction was once again encountered at 1270 m RKB while conducting

wiper trip prior to POOH and further reaming followed. A possible reason for the

development of obstruction between about 1200-1360 m RKB was thought to be the

presence of swelling clays. XRD analysis (Appendix B) confirms the occurence of swelling

clays around this section. After a wide consultation, a decision was made to pump a mixture

of 1.75 tonnes of sodium hexametaphosphate [(NaPO3)6] (a dispersant) and 10m3 of water

into the wellbore to help break the clay bonds. This proved successful. Eventually, the

wellbore was successfully reamed to the bottom using aerated water and foam, receiving full

circulation returns at the surface. A wiper trip was conducted and the drill string pulled out

of the hole before the 7" slotted liner and 2 plain liner pipes were RIH and squatted at the

bottom. This phase lasted 91 days.

23

3 Sampling and Analytical Methods

3.1 Sampling

Studies of various parameters in borehole geology require comprehensive and accurate

sampling. The information obtained depends on the samples quality. Consequently, a

systematic selection of samples is indispensable. Principally, this study involves sampling

of drill cuttings (chips), measurements at discrete depths and laboratory analyses. Drill

cuttings analysis is basically a visual method of semi-quantitatively describing the host rock

characteristics from drill cuttings obtained during a drilling operation. The analyses are

crucial in constraining the sub-surface conditions of the geothermal system (Reyes, 2000).

These encompass, but are not limited to, lithostratigraphy, hydrothermal alteration rank and

intensity, permeability, petrochemical characteristics and zonation patterns.

Unlike in the case of core samples, interpretation based on drill cuttings analysis is not

without a fair share of its challenges, including imprecise sample location as a consequence

of relatively large lag times. In other words, drill cuttings do not represent exact horizons,

but a limited depth range. The time taken by drill cuttings to get to the surface is dependent

on the type of circulation fluid, drillhole diameter, flow rate and circulation loss. In addition,

chances of collecting unrepresentative samples caused by sample mixing as a result of

drilling through soft formation are quite high. Further, grinding of cuttings to extremely fine

grains (almost powdery) by the drill bit leads to loss of essential geological features. No

cores were collected in well OW-922. Therefore, all the descriptions and interpretations in

the present study are based solely on the analysis of drill cuttings. Uncertainty in sample

depth does not substantially affect the conclusions reached in this study. Cuttings were

collected at 2 m interval, except in cases of partial circulation returns. In such cases, it was

carried out at 4 m interval. The sampling intervals along with total loss zone depths were

constantly recorded in the rig geology notebook.

3.2 Analytical methods

To achieve the specific objectives of this study, several analytical techniques have been

employed as discussed in a following section. The procedure and a summary of the analytical

results obtained are presented in different appendices.

3.2.1 Binocular microscope analysis

The main parameters described by binocular microscope analysis include colour of the rock,

grain size, rock fabrics, primary/original mineralogy, hydrothermal alteration rank and

intensity, rock type and features that point to the possible occurence of feed zones or

intrusions. Samples were scooped up using a petri dish and washed to enhance visibility and

reveal any obscured features. Binocular microscope analysis for this study was carried out

at Iceland Geosurvey (ÍSOR) petrography laboratory using an Olympus, SZX12 binocular

microscope. A detailed description of the lithology is presented in Appendix A.

24

3.2.2 Petrographic microscope analysis

With this analytical technique one can ascertain in much finer details the rock type and

hydrothermal alteration grade based on the optical properties of minerals. Hydrothermal

alteration intensity, paragenetic sequence and some other features not obviously discernible

under the binocular microscope can be described in detail. A total of 70 thin sections were

analysed at the University of Iceland petrographic laboratory, using, an Olympus BX51 and

Leitz Wetzlar petrographic microscopes with a magnification range of between 4x and 50x.

3.2.3 X-ray diffraction analysis

The X-ray diffraction (XRD) analysis technique is concerned primarily with structural

aspects of clay minerals and other minerals (e.g. zeolites, amphiboles, etc). The

diffractometry analysis is based on the principles of Bragg’s law. When a beam of X-rays is

passed through a particular mineral and certain geometric conditions are achieved (Yoshio

et al., 2011), the X-rays scattered from the crystalline sample can undergo constructive

interference, thus producing a diffraction peak. A diffraction pattern (made by the

diffractometer) records the X-ray intensity as a function of 2𝜃 (incident angle of the X-rays)

which is characteristic of a specific mineral. In this study, a total of 105 samples were

analysed exclusively for clay minerals at the KenGen XRD laboratory. A summary of the

XRD analytical results are presented in Appendix B.

3.2.4 Inductively Coupled Plasma-Optical Emission

Spectrometry (ICP-OES) analysis

The ICP-OES method has today become a valuable tool for quantitative elemental analysis.

The basic principle behind the method is that atoms of an analyte emit electromagnetic

radiation at certain characteristic wavelengths when atomized (Rollinson, 1995). The

wavelengths of the light emitted are then separated and measured in a spectrometer. This

yields an intensity measurement that can be easily converted to an elemental concentration

by comparison with calibrated reference standards (Rollinson, 1995). Actual analysis

involves spraying the sample solution into the core of an inductively coupled argon plasma,

generating temperatures of approximately 7,000-10,000 K. At such high temperatures, all

analyte species are atomized, ionized and thermally excited, and can then be detected and

quantified with an optical emission spectrometer (OES). The extremely high temperature

ensures that the sample matrix does not produce any interfering components that might

interfere with the analysis.

For this study, a total of thirty homogeneous samples from the study well were analysed for

major and trace elements using the Spectro Ciros 500 ICP-OES equipment at the Faculty of

Earth Sciences, University of Iceland. The major element oxides analysed are SiO2, Al2O3,

FeO, MnO, MgO, CaO, Na2O, K2O, TiO2 and P2O5. Trace elements analysed include, Ba,

Co, Cr, Cu, La, Ni, Rb, Sc, Sr, V, Y, Zn and Zr. Samples were analysed using the following

USGS internationally recognized standards and error margins; (1) Basalts: multiple

measurements of standard Basalt, Hawaiian Volcano Observatory (BHVO)-1 2σ (external

reproducibility). Analysis for Si, Al, Fe, Mn, Mg, Ca, Na and P was done with 2σ <1.8%. The error margin for trace elements in basalts ranged between 4-8%, (2) Rhyolites:

measurements of standard Rhyolite, Glass Mountain (RGM)-1. Elements were analysed with

25

2σ values as follows; Si and Mg ~ 1%, Ti <2%, Al, Ca, Fe, Na and K <4-6%, Mn and P

<8%. The error margin for trace elements in rhyolites is the same as for basalt. A detailed

description of the ICP-OES procedure for sample preparation, instrument set-up and data

processing, is presented in Appendix C. Full analytical results are listed in Table 5.

3.2.5 Electron microprobe analysis

Electron microprobe (EMP) analysis is routinely used for characterizing the compositions

and compositional variations of minerals within rock samples. The basic theory of EMP

analysis relies on the production of electrons and X-rays during the interaction of the incident

electron beam with the specimen (Schiffman and Roeske, 2014). Consequently, with the

electron beam focused on a small area on a sample surface, characteristic X-rays are

generated for qualitative or quantitative analysis at micrometer-scale spatial resolution. The

volume of the electron-matter interaction varies directly with the accelerating potential of

the electron beam (Schiffman and Roeske, 2014) and inversely with the mean density of the

sample. Two key processes, namely, the generation of backscattered electrons and the

generation of characteristic X-rays, are responsible for producing the main signals utilized

in EMP analysis.

A total of 10 samples from the four wells (OW-905A, 910, 917 and 922) were selected for

analysis of secondary minerals of interest. These represent four samples containing epidote

and six samples containing chlorite. The cuttings were subsequently moulded in epoxy,

polished, carbon coated and analyzed using a JEOL JXA-8230 electron microprobe located

at the Institute of Earth Sciences, University of Iceland. Elements analysed and the

conditions under which the probe was run are described in section 4.9. Prior to quantitative

analysis with the electron microprobe, the minerals of interest, namely, chlorite and epidote,

were first studied with a Hitachi TM 3000 Scanning Electron Microscope (SEM). Results

for EMP analysis for chlorite and epidote are presented in Appendix D.

27

4 Results

4.1 Lithology of well OW-922

Knowledge of the sequence, nature and architecture of the stratigraphy is of paramount

importance for understanding the geothermal system around well OW-922. Analyses based

on binocular and, petrographic observations and ICP-OES have revealed eight types of

lithologies intersected by well OW-922. These comprise pyroclastics, tuffs, rhyolite, basalt,

basaltic trachyandesite, trachyte, trachydacite and trachyandesite. No intrusions were

intersected by the well. A description of all lithological units encountered in the well are

discussed below and presented in Figure 14. However, a detailed description of the various

units and intervals of occurence along the well profile is presented in Appendix A.

4.1.1 Pyroclastics

The pyroclastic formation is a member of the Upper Olkaria Volcanics according to the OVC

stratigraphical classification scheme. The unit occupies the uppermost 302 m of the well

column.The formation generally varies in colour from light grey to yellowish-brown,

comparatively heterogeneous and indurated deposits dominated by ash-sized particles,

volcanic glass, scoria, highly vesicular pumiceous and tuffaceous fragments. It is oxidized

in all cases. The yellowish-brown unit are pyroclastics from Olkaria, whereas the light grey

to grey in colour counterparts are pyroclastics derived from Longonot volcano. Longonot

pyroclastics extend from the surface to ~ 96 m depth, whereas, Olkaria pyroclastics, derived

from the early post-caldera activity (Op2) are observed between ~ 190 and 302 m depth.

4.1.2 Tuffs

Tuffs in the well occur predominantly as interbeds and intercalations between lava flows,

marking different eruption episodes. The units range in thickness between 2 and 52 m. Two

distinct tuff varieties have been recognized in the well, lithic and glassy (vitric) types. The

former is the dominant, varying in colour from light grey, grey to brown and, occasionally,

pale green. The latter is due to alteration. Trachytic and basaltic fragments are the most

common crystalline fragments, and are embedded within the lithic matrix. The phenocryst

phases observed in optical petrography are quartz, sanidine and augite. Moderate to high

vesicularity is a common feature of these tuffs, with majority of the vesicles being irregular

in shape. Vesicles are filled by alteration minerals, primarily zeolites and chalcedony. Lithic

tuff was noted occuring between 302-350 m, 362-364 m, 408-422 m, 476-498 m, 564-582

m, 726-744 m, 952-966 m, 1440-1600 m and 1600-1642 m. Vitric tuff, on the other hand,

was only observed between 1260-1312 m. The vitric tuff is white in colour, and has a glassy

matrix as observed under the petrographic microscope, with minor sanidine and quartz

phenocrysts. Generally, the tuffs display moderate to high intensity of alteration, particularly

in the basal rubble (contacts).

28

4.1.3 Rhyolite

The rhyolite encountered in this well is of granular type and ranges in colour from light grey

to grey to pink. It is massive, holocrystalline, and generally sparsely to moderately phyric.

Also, it is hard, probably due to intense silicification. The matrix has commonly quartz

porphyries and subordinate sanidine phenocrysts. Small mafic mineral grains are

disseminated in the aphanitic groundmass and have been proved through optical petrography

to be riebeckite and oxides, probably magnetite. The formation is not extensive in the study

well as compared to that observed in wells drilled along and within the proposed caldera

structure. It is restricted to depths between 1046-1150 m, 1214-1224 m, and 2370-2592 m.

Based on petrochemical data, the rhyolite is silica-rich with SiO2 content ranging between

69 and 73 wt.%.

4.1.4 Basalt

The formation ranges in colour from dark grey to brown, with the brown varieties depicting

high intensity of hydrothermal alteration. The dark grey basalt is fine grained and

holocrystalline as observed between 498-504 m, 1198-1214 m and 1642-1654 m. In addition,

it is dense and the predominant phenocryst phases are plagioclase and clinopyroxene

(notably augite) as well as a large abundance of opaque oxides, mainly titanomagnetite.

Olivine phenocrysts are rare and their prescence is disclosed by crystal outlines of the

alteration products, e.g. calcite and chlorite. Plagioclase laths are also widespread in the

holocrystalline matrix and outweigh clinopyroxene in all cases. On the other hand, vigorous

alteration has completely obliterated the primary mineral phases and original texture in the

brown coloured basalt. The latter occurs between 1150-1198 m. Nevertheless, vesicles are

observed largely filled with calcite and variably green clays in both cases.

4.1.5 Basaltic trachyandesite

This rock type of intermediate composition is scarce in the well, and generally along the

EARS system based on petrochemical data from previous studies (e.g. Peccerilo et al., 2007).

It was identified between 1654 and 1700 m through petrochemical analysis. The rock is

mostly brown to grey (probably as a result of intense alteration), though it occasionally

appears light grey (e.g. between 1662-1666 m and 1680-1682 m). It is aphyric to slightly

phyric with minor plagioclase phenocrysts. Other phenocryst phases observed are

amphiboles and clinopyroxenes. The latter occur characteristically as slender green needles.

Mafic silicates in most cases (~ 80%) display more or less complete replacement by brown

clays.

4.1.6 Trachyandesite

Like basaltic trachyandesite, this rock type is also scarce in this well and was confused for

trachyte during the initial binocular analysis. The unit was identified through petrochemical

analysis as occuring only between 2780-2784 m. The rock has a composition intermediate

between trachyte and andesite and represents less fractionated magma. The notable

phenocryst phases observed in thin section are plagioclase, sanidine and pyroxenes, mainly

aegerine-augite.

29

4.1.7 Trachydacite

This unit was identified between 1700-1730 m, 2186-2210 m, 2326-2370 m and 2592-2780

m through petrochemical analysis. The rock has a composition intermediate between

trachyte and dacite. The colour varies from pale grey to grey, to pink. It is aphyric to poorly

phyric, and contains clinopyroxene and sanidine as the principal phenocrysts. Vesicles are

rather scarce, but where present are filled by calcite, chlorite and secondary quartz.

4.1.8 Trachyte

This is the dominant lithology in the well and the main sequence is ~ 1200 m thick, as

observed between 1700-2900 m. It is the reservoir rock of the geothermal system around

well OW-922. This unit is predominantly grey, though colour ranging from pale green to

brown is also noticed. In most cases, the unit is phyric with sanidine, subordinate feldspars

and quartz phenocrysts as well as specks of mafic minerals such as pyroxenes and

amphiboles. The pyroxenes are in many cases aegerine-augite while the dominant amphibole

is dark blue riebeckite, recognised through petrographic analysis. Occasionally, the

formation is observed mixed with basaltic and tuffaceous fragments (e.g. between 1248-

1260 m and 2156-2292 m), suspected to have fallen from above during drilling. Trachytic

texture with flow banding is a characteristic feature of these trachytes with sanidine laths

aligned in a particular direction. In addition, elongated vesicles are abundant, a display of

the direction of lava flow.

4.2 Hydrothermal alteration

Hydrothermal alteration is a metasomatic process involving recrystallization of rock

forming/primary minerals and deposition of secondary minerals in voids, to a new authigenic

mineral assemblage due to interaction with thermal fluids. The new mineral assemblage is

more stable, or at least metastable, at a specific temperature, pressure and fluid composition

range within a particular geothermal system. The stability of the hydrothermal minerals is

determined by their solubility and the rock dissolution process. Hydrothermal alteration

involves interaction of the hot aqueous fluids with the host rock (a process popularly known

as water-rock interaction), resulting in a change both in the composition of the fluids and

the rocks (Brown, 1978). Circulation of thermal fluids through the geothermal system is a

consequence of the heat generated by magma or hot crustal rocks.

During ascent to the surface along permeable structures, the thermal fluids circulating within

the geothermal system directly deposit dissolved mineral constituents in vesicles, vugs or

veins. Correspondingly, the thermal fluids dissolve ions from the host rocks, consequently

forming secondary minerals. By observing alteration overprints or the sequences of

hydrothermal minerals deposited in the veins or vugs, it is possible to unravel the prevailing

reservoir temperature, permeability and fluid composition (Reyes, 2000). According to

Mielke et al. (2015), hydrothermal alteration is defined by both its grade and intensity. The

term grade implies a definite assemblage of hydrothermal minerals present. Intensity of

alteration, on the other hand, defines whether the alteration is incipient, partial, or complete.

Accordingly, alteration is described using qualitative terminology such as low, medium, or

high, depending on if a primary constituent is incipiently, partly, or completely altered in

that order (Reyes, 2000).

30

Formation of hydrothermal minerals is generally dependent on the temperature, pressure,

reservoir permeability, protolith, fluid composition (especially pH), and duration of fluid-

rock interactions (Brown, 1978). Although pressure has little direct effect on hydrothermal

alteration, an important aspect is that it controls the depth at which boiling occurs, and thus

separation of vapour and gases. Some of these factors listed are so intimately interconnected

that, at first sight, it is often impossible to separate one from another (Brown, 1978). Products

of hydrothermal activity in the geothermal system in the vicinty of well OW-922 occur both

as replacement of unstable primary components or glassy matrix of the rocks (Table 3) and

as open space fillings in veins, fractures, and vesicles.

Table 3: Alteration of primary minerals and order of replacement in well OW-922

(adapted from Brown, 1984).

Order of replacement

Su

sceptib

ility

Primary phases Alteration products

Volcanic glass

Olivine

Plagioclase

Pyroxenes, amphiboles

Sanidine

Fe-Ti oxides (opaques)

Clays, calcite

Clays, chlorite, calcite, quartz

Albite, clays, zeolites, calcite

Clays

Adularia, clays, calcite

Titanite, pyrite, haematite

4.2.1 Distribution of hydrothermal alteration minerals

Hydrothermal alteration minerals in well OW-922 were observed to occur both as

replacement of primary minerals, and as deposition as veins and as fracture and vesicle

fillings. The distribution of alteration minerals in well OW-922 was determined with

binocular, petrographic and XRD analyses. The main hydrothermal minerals encountered in

the well include: zeolites (mesolite, scolecite and wairakite), chalcedony, haematite, titanite,

calcite, albite, pyrite, clays (smectite, illite, chlorite and mixed-layer clays (MLCs)), quartz

and epidote. This is presented in Figure 14 and discussed further below.

Zeolites are hydrated aluminosilicate minerals structurally characterized by a framework of

linked tetrahedra, each consisting of four oxygen atoms surrounding a silicon or aluminium

cation. This three-dimensional network has open cavities in the forms of channels and cages

(Wise, 2005), which are occupied by water molecules and non-framework cations.

Occurence of zeolites in geothermal environments is usually typified by three different

fabrics, namely, fracture and vesicle fillings, as replacement products of relict feldspars, and,

as recrystallization products of siliceous or glassy groundmass of volcanic rocks (Wise,

2005). Most zeolites, with the notable exception of wairakite, are low-temperature minerals

which, are thermodynamically unstable at temperatures >110oC.

Unlike in most wells in the Olkaria geothermal system, the occurence of zeolites in this well

is restricted only to two low-temperature types, mesolite, a high-silica zeolite, and scolecite,

a low-silica zeolite. Both were observed largely as vesicle infillings in tuffs between 302-

350 m, 362-364 m and 408-422 m, and in trachyte between 364-408 m. Mesolite was

identified by the occurence of white, thin, needle-like crystals, which radiate away from the

base, forming spiky ends. By contrast, scolecite was in most cases colourless and formed

clusters radiating from a central point of growth.

31

Wairakite, a calcium-rich zeolite, was first observed in a vesicular tuff between 494-496 m.

It occurs in association with hydrothermal quartz and more rarely with calcite. It is

characterizd by a colourless to murky white colour with aggregates of subhedral crystals,

coating fractured surfaces. It is also identified filling out the abundant pore spaces in

pumiceous fragments at 574-578 m. The occurence is nonetheless sporadic. The first

appearence of wairakite at 494 m in well OW-922 signifies a temperature of about 200oC

and it seems to be the only zeolite stable above that particular temperature (Kristmanndóttir,

1979).

32

Figure 14: Lithology, location of feed zones, distribution of alteration minerals and

alteration zones of well OW-922.

33

Chalcedony is a cryptocrystalline variety of quartz, characterized by a translucent white to

grey to light blue, and in some instances shades of brown. The mineral was first observed at

412 m as a white thin-layered lining in a vesicle. It is also found intermittently and occurs in

the deeper sections of the well, for example between 2576-2578 m. Chalcedony forms at

temperatures from 150oC to as high as about 180oC (Kristmanndóttir, 1979), but gradually

recrystallizes to hydrothermal quartz with increasing temperature (e.g. between 2114-2116

m).

Haematite is a low-temperature Fe-oxide whose occurence is associated with the incursion

of cold groundwater and other oxidizing conditions in a geothermal system (Brown, 1978).

The mineral was observed by binocular analysis as a silver-grey, brown to reddish brown in

colour. It was noted in moderate amounts between 1192-1200 m, and occurs irregularly from

2342 to 2982 m.

Titanite is recognized as white coloured mineral, usually as a replacement of the mafic

silicates. The mineral is very common in this well and was first observed between 422-476

m, occuring sporadically to the well bottom. In petrographic analysis, titanite is identified

by its high relief, brown to dark brown colour and shows dusty-cloud discoloration formed

by the alteration of the ferromagnesian minerals.

Calcite is ubiquitous in the well and occurs in all the hydrothermal mineral alteration zones.

The occurence is particularly prevalent between 1120-1730 m and 2210-2990 m, where it is

rated at a scale of between 2 and 3 (Figure 14). The latter indicates high abundance, whereas,

the former signifies moderate abundance. The mineral is noted as a replacement of primary

forming minerals, especially plagioclase and pyroxene phenocrysts, and discrete silicic

volcanic glass. This process inevitably requires introduction of CO32−ions from the

geothermal fluids (Steiner, 1977). In addition, calcite is observed as precipitates in veins and

vesicles (e.g. between 520-522 m, 1600-1646 m) and as platy calcite (e.g. between 1646-

1648 m and 1944-1946 m), possibly indicating deposition through mixing of fluids with

different compositions. However, in some rare cases, it is noted as a complete groundmass

replacement. In active geothermal systems, calcite is usually regarded to be stable over a

broad temperature, ranging from ~ 50oC to about 300oC (Simmons and Christenson, 1994).

Pyrite is a hydrothermal mineral whose presence in significant amounts has traditionally

been associated with permeable zones in most geothermal systems. It was first observed at

302 m in a tuff. Like calcite, pyrite occurs pervasively to the well bottom. The occurence

varies between 1 to 3, implying low and high abundance, in that order. In binocular analysis,

the mineral was recognized by euhedral cubic crystals with a shiny brass yellow lustre,

predominantly disseminated and, in other instances, embedded in the rock matrix. On rare

occasions, for example between 1312-1420 m, the mineral is seen filling veinlets. The cubic

crystal nature was equally applied while distinguishing pyrite from the other opaque

minerals in petrographic analysis. According to Steiner (1977), pyrite is a product of the

replacement of magnetite (equation i) or result from the attack of FeO in ferromagnesian

minerals with H2S gas (equation ii); such processes are known to occur in other geothermal

systems.

(i) Fe3 O4 + 6H2S ↔ 3FeS2 + 4H2O + 2H2

(ii) FeO + 2HsS ↔ FeS2 + H2O + H2

34

Furthermore, the Fe in the pyrite is suggested to be possibly introduced by the geothermal

fluids, which commonly contain minute amounts of Fe (Steiner, 1977). An alternative

interpretation is that the Fe may be leached from neighboring rocks and transported perhaps

only for a short distance before being converted to pyrite. Pyrite also occurs over a broad

temperature, ranging from ~ 50oC to about 300oC (Brown, 1978).

Hydrothermal quartz is observed mainly as an open-space filling mineral in veinlets,

vesicles, vugs and interstices of fractured rocks. It is generally colourless to occasionally

white coloured, hexagonal shaped, commonly found as subhedral to euhedral crystals that

rarely exceed ~ 2 mm in length. The mineral was first noted between 476-478 m filling

vesicles in a tuff, designating formation temperature in excess of 180oC (Reyes, 2000). The

occurence is however, sporadic to the well bottom. In many cases, it is found in association

with wairakite, pyrite and titanite. In thin section, quartz was easily identified by the

undulating extinction, lack of cleavage and low interference colours. Petrographic evidence

unequivocally shows that quartz to a greater extent precipitated or was introduced by

geothermal fluids as opposed to crystallised from SiO2 released during alteration of silicic

glass and/or feldspar phenocrysts and ferromagnesian minerals.

Albite. The term albitisation in this context is reserved for the replacement of primary K-

feldspar (specifically sanidine in this case) and plagioclase phenocrysts by hydrothermal

albite. The upper limit of this mineral in the well is at 802 m and it occurs irregularly but

disappears totally below 952 m depth. The mineral was identified during binocular

investigation by its cloudy white and striated surfaces. The mineral grains vary largely

between anhedral to subhedral in all cases. Albite is formed at a temperature of 180oC or

higher (Brown, 1978), and its presence implies zones of low permeability.

Epidote was first noted at 1542 m through petrographic analysis from its high relief, pale

green to pale yellow and greenish brown pleochroism, in addition to parallel extinction. The

interference colors are distinctively bright (second order). Generally, epidote is sporadically

distributed in this well and is found more commonly as a replacement mineral of the primary

feldspars (plagioclase and sanidines) and to a lesser extent the pyroxenes. It occurs at discrete

depths and disappears completely below 2784 m. The minerals that were commonly found

in association with epidote include mixed-layer clays (MLCs), secondary quartz, calcite,

pyrite and titanite. In drill cuttings analysis, epidote was identified by its conspicous yellow

to greenish yellow colour. Pure crystalline epidote is rarely present and in most instances

(e.g. 2234-2236 m and 2570-2580 m) the mineral was observed only as a faint coloration.

In most geothermal systems, the appearence of epidote indicates temperatures equal to or

above 240oC (Bird and Spieler, 2004).

Clay minerals

The ubiquitous nature of clay minerals as a function of physico-chemical environment of

crystallization has enabled their application as markers of paleo-conditions in active

geothermal systems. Geothermal systems are especially fruitful places to study clay minerals

genesis since these environments serve as natural laboratories where fluid-rock interactions

occur under measurable conditions (Steiner, 1968). The distribution of clay minerals in

active geothermal fields depends on the ability of the fluid to approach equilibrium with host

rocks at any scale during the hydrothermal processes. Clay mineral transformations are

governed by the Ostwald Step Rule (Moore and Reynolds, 1997). According to the rule, clay

mineral paragenesis evolves through the formation of sucessive metastable phyllosilicates

35

that progress towards the state of stable chemical and textural equilibrium. The minerals

have proved to be particularly useful for temperature estimations (mineral geothermometers)

based either on modifications of crystal structure (crystallinity, interstratification, order-

disorder, etc) or modifications of crystal chemistry (layer charge, substitutions, etc) of the

crystallites (Moore and Reynolds, 1997). Clay minerals in well OW-922 occur largely as

replacement products of volcanic glass, feldspars, ferromagnesian minerals and rarely as

depositional products in veins and vesicles. In the latter cases, the minerals are found in

association mostly with hydrothermal quartz. Identification of the clay minerals was

primarily based on the position of d-spacings determined with XRD after air-drying,

saturation with ethylene glycol and heat treatment. X-ray diffraction analysis of drill cuttings

from well OW-922 revealed four distinct minerals of the phyllosilicate group as discussed

below. Representative X-ray data for clay minerals observed in this well are compiled in

Appendix B, and selected X-ray diffractograms shown in Figures 15 and 18.

Smectite

Smectite, also known as ‘‘swelling clay” is a low-temperature clay alteration mineral and

was first noted at 338 m depth through XRD analysis. It forms at comparatively low

temperatures, up to 120oC, though generally not exceeding about 200oC (Brown, 1984). In

the study well, the occurrence of smectite is widespread, occurring at shallow depths of the

well (338-450 m) and extending to depths between 1138-1166 m. The mineral is noted

largely replacing volcanic glass and ferromagnesian minerals. In binocular analysis, the

mineral was identified by its extremely fine-grained and poorly crystallized nature,

particularly in the uppermost 450 m. The color varies from light to dark green to brown.

According to Steiner (1968), smectite is commonly the first clay mineral phase to replace

volcanic glass and likely represents a metastable phase of reaction progress involving glass

and hydrothermal fluids. Diffraction patterns of the untreated air-dried samples show broad

d(001) peaks ranging from 12.4 Å to 15.0 Å (Figure 15). The basal diffraction peak is quite

sensitive to ethylene-glycol treatment and invariably widens to higher spacing, ranging from

16.9 Å to 17.5 Å. However, the peak irreversibly collapses upon heating to about 9.97 Å

(Figure 15), corresponding to the theoretical spacing for ideal smectite. The observed

changes of the first-order, basal reflection, d-spacing between peaks is caused by variation

of interlayer water molecules.

Laird (2006), observed that smectites swell by absorbing water or polar organic solvents

between smectite quasicrystals and/or between the individual layers within quasicrystals.

Such swelling occurs when the clay is dispersed in a solvent, or when the clay is in direct

contact with an atmosphere having a high vapor pressure of the solvent. Smectite is

associated with a secondary mineral assemblage consisting principally of pyrite, zeolites and

quartz. In many geothermal systems (Steiner, 1968), smectite has been noted gradually

transforming to illite in response to increasing temperature or depth through a series of

progressively more ordered smectite/illite interlayer structure. In contrast, cases of illite

converting back to smectite have equally been observed in other geothermal fields, e.g. in

Wairakei (Steiner, 1968), particularly where there is an inversion of temperature.

36

Mixed-Layered Clays (MLCs)

The term MLCs is synonymous with interlayered or interstratified clay minerals. Essentially,

MLCs represent clay minerals derived from two or more kinds of intergrown layers as

opposed to physical mixtures (Moore and Reynolds, 1997). Most often, MLCs contain

smectite as a swelling/expanding component. The interlayering is defined by three particular

types, i.e. regular (ordered), random or segregated (Moore and Reynolds, 1997). In well

OW-922, these clays mark the transitional stage from swelling to non-swelling varieties.

They are noted as the most predominant clay mineral phases, occurring regularly to the

bottom of the well. MLCs apparently can form in many possible combinations of different

layers. However, the common varieties in this case are smectite/illite, smectite/chlorite and

the three component type, smectite/illite/chlorite. The latter is the most abundant and

widespread form, occurring to the well bottom. It is characterised by a d-spacing of 15.4-

12.2 Å for untreated samples and 17.8-16.5 Å for ethylene glycol solvated samples. The

expanded layers collapse to about 9.98 Å upon heating. The chlorite component is indicated

by the 7 Å peak which collapses totally as a result of oven heating. MLCs were not identified

in binocular analysis but were recognized by the relatively more yellowish to green colour,

stronger pleochroism and brighter birefringence colours (Figure 16) relative to chlorite in

optical petrography. MLCs were first noted at 1038 m through XRD analysis (Appendix B).

EMP analysis (analyses not shown) show relatively higher K2O (7.7-9.8 wt.%) for MLCs

compared to chlorite (0.12-0.62 wt.%). These clays are normally stable at temperatures ≤

220oC.

OW-922 358-360 m

File: 358-360H_1.raw - Type: 2Th/Th locked - Start: 2.000 ° - End: 15.000 ° - Step: 0.020 ° - Step time: 1. s - Temp.: 25 °C (Room) - Time Started: 0 s - 2-Theta: 2.000 ° - Theta: 1.000 ° - Phi: 0.00 ° - Aux1: 0.0 - Aux

File: 358-360G_1.raw - Type: 2Th/Th locked - Start: 2.000 ° - End: 15.000 ° - Step: 0.020 ° - Step time: 1. s - Temp.: 25 °C (Room) - Time Started: 0 s - 2-Theta: 2.000 ° - Theta: 1.000 ° - Phi: 0.00 ° - Aux1: 0.0 - Aux

File: 358-360A_1.raw - Type: 2Th/Th locked - Start: 2.000 ° - End: 15.000 ° - Step: 0.020 ° - Step time: 1. s - Temp.: 25 °C (Room) - Time Started: 0 s - 2-Theta: 2.000 ° - Theta: 1.000 ° - Phi: 0.00 ° - Aux1: 0.0 - Aux

Lin

(C

ou

nts

)

0

100

200

300

400

500

2-Theta - Scale

2 3 4 5 6 7 8 9 10 11 12 13 14 15

d=

16,7

6951

d=

12,4

8563

d=

9,9

7703

d=

6,4

6542

d=

8,4

1491

Smectite

Figure 15: X-ray diffractogram of air dried, glycolated and heated sample of smectite.

37

Illite

Illite was first noted at 474 m through XRD analysis. The mineral occurs at temperatures of

approximately 200oC or more. It is observed to form mostly from the alteration of feldspars,

even though precipitates in veins and vesicles are also noted in small quantities. The mineral

is associated with a mixture of secondary minerals including variable but generally trace

amounts of hydrothermal quartz, chlorite, and pyrite. In thin section, illite is colourless to

yellowish brown, with the texture varying between fine to coarse. In many cases (e.g.

between 1238-1240 m), it was observed as matted flakes or fibrous aggregates. In XRD

analysis, the diffraction patterns for the untreated sample show a first order basal reflections

at ~ 10 Å and do not show significant departure from the 10 Å integral series upon ethylene

glycol solvation and heating.

Chlorite

Chlorite is a common clay mineral in the Olkaria geothermal system whose occurence

indicates reservoir temperatures exceeding 220o C (Kristmanndóttir, 1979). In well OW-922,

chlorite formed as a replacement product of ferromagnesian minerals (mostly pyroxenes)

and feldspars (sanidine and plagioclase), and also as densely packed aggregates filling

vesicles. It occurs preferentially in basalts and trachyte and is associated with mineral

assemblages of quartz, calcite and pyrite, and rarely, epidote. According to the XRD results,

pure chlorite is seldom found as an individual phase, but in many cases the mixed-layer clays

has chlorite structure layers along with smectite and illite layers (Appendix B).The colour

varies widely from green to bluish green as observed in binocular investigation. In thin

section, the mineral is fibrous and displays light to dark green colours, weakly pleochroic

from yellow green to green (Figure 17). In addition, it has low interference colours (first

order gray). Occasionally, the mineral occurs in minute amounts forming spherulites.

Figure 16: Cross polarized image of mixed-layer clay depicting high

interference colours at a depth of 2782 m.

38

In XRD analysis, the spacings of the first order d(001) basal reflections at ~12.4-14.7 Å are

not affected by air drying, glycol solvation or heat treatment. This peak however, is obscured

in most samples in this well (e.g. Figure 18), leaving the second order reflections d(002)

stable at ~7.0-7.1 Å for untreated and glycol treatment as the only conspicuous peaks. The

latter, totally disappears upon oven heating though. Previous studies (e.g. Otieno and Kubai,

2013) have termed this chlorite as an unstable variety and it is the most common chlorite

type in many wells in the Olkaria sub-fields. A widely quoted hypothesis by Moore and

Reynolds (1997) suggests this kind of chlorite to be Fe-rich. EMP analysis results for

chlorite, discussed in section 4.9, are fully in agreement with this observation. Based on their

argument, the Fe-rich component in the chlorite is introduced by prolonged interaction

between the adjacent host rocks and the altering solution.

Figure 17: A plane polarized image of fibrous chlorite filling a vesicle

at 1826 m.

39

4.2.2 Hydrothermal alteration zonation

Many high-temperature geothermal systems of the world are, divisible into different zones

based on changes in alteration assemblages. Different alteration minerals become stable or

metastable within a particular temperature range and with certain fluid compositions. Thus,

by mapping the mineral sequences as well as the alteration minerals, it is possible to deduce

the thermal evolution and conditions of a specific geothermal system (e.g. Mortensen et al.,

2014). Alteration zones are characterised by an abundance of particular minerals

corresponding to increased temperature and depth. The top of each alteration zone is defined

by the depth of the first appearance of a mineral. Four alteration zones, namely zeolite-

smectite, illite-chlorite, mixed-layer clay zone and epidote-mixed-layer clay zone, were

distinguished below the unaltered zone based on optical petrography and binocular

observations. Representative alteration mineral assemblages of each zone are listed as well.

Identification of the different clay types in this well relied largely on XRD results (Figure

14), and to a lesser extent on electron microprobe analysis.

0-190 m depth: Unaltered zone: This is the lowest grade alteration zone in this well. It is

composed of relatively fresh rocks, mainly pyroclastics with no alteration signature related

to geothermal activity, an indication that the temperatures have been < 50oC. Volcanic glass

constitutes the bulk of the rocks and are conspicuous both in binocular and petrographic

observation. The unaltered zone is characterised primarily by oxidation, which supposedly

OW-922 1860-1862 m

File: 1860-1862H_1.raw - Type: 2Th/Th locked - Start: 2.000 ° - End: 15.000 ° - Step: 0.020 ° - Step time: 1. s - Temp.: 25 °C (Room) - Time Started: 0 s - 2-Theta: 2.000 ° - Theta: 1.000 ° - Phi: 0.00 ° - Aux1: 0.0 - A

File: 1860-1862G_1.raw - Type: 2Th/Th locked - Start: 2.000 ° - End: 15.000 ° - Step: 0.020 ° - Step time: 1. s - Temp.: 25 °C (Room) - Time Started: 0 s - 2-Theta: 2.000 ° - Theta: 1.000 ° - Phi: 0.00 ° - Aux1: 0.0 - A

File: 1860-1862A_1.raw - Type: 2Th/Th locked - Start: 2.000 ° - End: 15.000 ° - Step: 0.020 ° - Step time: 1. s - Temp.: 25 °C (Room) - Time Started: 0 s - 2-Theta: 2.000 ° - Theta: 1.000 ° - Phi: 0.00 ° - Aux1: 0.0 - A

Lin

(C

ou

nts

)

0

100

200

300

400

500

600

700

800

2-Theta - Scale

2.1 3 4 5 6 7 8 9 10 11 12 13 14 15

d=

7,1

5824

d=

6,3

9214

d=

10,0

1907

d=

14,1

8330

Illite and Unst. chlorite

Figure 18: X-ray diffractogram of illite and unstable chlorite showing obscured

peak of 14.18 Å for the three treatments and 7.15 Å peak for air dried and glycol

solvation. The latter peak totally collapses upon heating.

40

is related to the interaction between the host rocks and the comparatively oxygen-rich

descending cold groundwater. Total loss of circulation between 96-190 m (Appendix A)

made it impossible to locate precisely where this zone terminates within the specified depth

range.

190-474 m depth: Zeolite-smectite zone. The alteration mineral assemblages in this zone are

characterised by low-temperature zeolites, smectite clay and trace quantities of pyrite and

calcite. Other notable minerals include chalcedony and titanite. The low-temperature zeolites

present in this zone are mesolite and scolecite. Smectite was first recognized by XRD

analysis at 338 m (Appendix B), suggesting that the temperatures in this zone are < 180-

200oC.

474-1038 m depth: Illite-chlorite zone. The concurrent appearence of illite and chlorite at

474 m defines the top boundary of this zone, signifying that the formation temperature is

gradually increasing with depth and is in excess of 200oC. In conjunction with this, the

zeolites (mesolite and scolecite) show indications of instability and completely disappear at

482 m. Quartz and wairakite appear and are first identified at 476 m and 494m, respectively.

According to XRD analysis, the chlorite is an unstable variety. Other alteration minerals

found in this zone are chalcedony, titanite, calcite, albite and pyrite. Calcite is ever-present

as open-space and matrix fillings, while pyrite is prevalent as patchy disseminations in the

rock groundmass.

1038-1542 m depth: Mixed layer clay (MLC) zone. The characteristic feature of this zone is

the first appearence of MLCs, in the form of smectite/chlorite at 1038 m. Based on the

discussion under MLCs, temperatures in this zone are ≤ 220oC. Common assemblages

continue to include pyrite, calcite, haematite and titanite.

1542-2990 m depth: Epidote-Mixed layer clay zone. Epidote first occurs at 1542 m and

defines the top of this zone. It is the largest zone extending to the well bottom. The first

appearence of epidote indicates that the formation temperature has been at or higher than

240oC. The occurence of this mineral is very sparse in comparison to MLCs. Epidote is

observed to occur until about 2784 m, after which it totally disappears to the bottom of the

well. In addition to MLCs, epidote, quartz, titanite, chalcedony, haematite and pyrite, calcite

is by far the dominant mineral phase. The mineral is observed overprinting epidote at 2782

m.

4.2.3 Distribution of alteration mineral zones in the Domes area

Alteration mineral zones are basically the result of diagnostic alteration minerals present

within a particular geothermal system, and can give reliable indications of the reservoir

temperature. The formation of hydrothermal minerals, as discussed above is not solely a

function of temperature, but also fluid composition, initial composition of the rock system,

permeability, etc (Brown, 1978). In Figure 19, the distribution of alteration mineral zones in

the study well is compared with results of previous studies (e.g. Musonye, 2015; Mwangi,

2012). From this figure, it is evident that zones indicative of high temperatures, e.g. illite-

chlorite-epidote and chlorite-epidote-actinolite, marked by the first appearence of epidote

and actinolite, respectively are present at relatively shallow depths ( between 1280-1440

masl and 800-1240 masl, respectively) in wells OW-910, 916 and 914. The same zones

appear at relatively greater depths in well OW-917. However, in well OW-922, the two zones

are conspicuously absent, even though it is known from hydrothermal mineralisation that the

41

first appearence of epidote in this well occur at even much deeper depth, i.e. at 458 masl.

Observations made for the distribution of alteration mineral zones across a section of the

Domes field positively correlate with the model for measured formation temperature

distribution across the same field (Figure 35). High formation temperatures are observed at

comparatively shallow depths in wells OW-910, 916 and 914, whereas, at greater depths in

well OW-917. In contrast, well OW- 922 is generally characterised by low temperatures.

Figure 19: Distribution of alteration mineral zones across a

section of the Domes area, along the traverse A-B.

42

4.3 Feed zones

Feed zones are basically water-saturated regions in the sub-surface which produce an

economically feasible quantity of water to a well. The primary goal of drilling a geothermal

well is to intersect as many permeable and hot feed zones as possible, below the production

casing, so as to maximise the well productivity. For locating feed points during drilling, one

can rely on a combination of observations, including, circulation losses or gains, intrusive

boundaries, lithological contacts, changes in alteration mineralogy patterns, appearance of

highly fractured formations, changes in circulating fluid temperatures, as well as temperature

recovery logs. In the Domes area, the location of feed points from previous studies (e.g.

Mwangi, 2012; Ronoh, 2012; Lagat, 2004) have been inferred primarily from lithological

boundaries, highly fractured lavas, sudden increase in the rate of penetration and variation

in alteration patterns.

In drill cuttings analysis of the study well, fluid “feeder” zones below the production casing

were deduced from lithological interfaces, alteration intensities and drilling parameters (i.e.

rate of penetration (ROP)). Observations of circulation losses have equally played some role,

but not to a large extent since most of the loss zones encountered range between 2 and 14 m,

and in most cases are principally attributed to a change in stand-pipe pressure during POOH

to change a drill bit. Temperature and pressure recovery were determined by four heat-up

profiles, i.e. after 9 hours and, 12, 21 and 49 days. Pre-injection temperature and pressure

profiles were taken prior to the recovery tests. However, it is worth noting that no injection

tests were conducted since the well got filled up with water during injection, which argues

against the existence of any major feed zone(s). As a consequence, no injectivity index was

measured for this well. Another key point to highlight is that the temperature recovery over

a period of only 49 days is not sufficient for the geothermal system near the well to reach

stable conditions following the long drilling time (Figure 13).

Step pumping was conducted between 19th September and 6th November, 2014. The first

pumping rate was set at 1000 l/min, and then subsequently increased to 1300 l/min, 1600

l/min and finally 1900 l/min. Traditionally, this is the pumping rate, which has been adopted

during step pumping of the Olkaria wells. During all pumping rates, temperature and

pressure measurements were carried out at 100 m interval from the surface to 1200 m, which

is the production casing depth. Afterwards, the interval was reduced to 50 m to the well

bottom. These intervals are constant for all wells. However, the change from 100 to 50 m is

subject to the casing depth, which is quite variable from one well to another. Pressure logs

measured in this well (Figure 20) display no intersection of the various pressure recovery

profiles, commonly referred to as the "pivot point".

43

From the temperature recovery logs (Figure 21), it is observed that temperature increases

from the surface to about 2250 m at an average rate of 4.5o C/100 m depth. This is followed

by a more or less constant temperature between 2300-2500 m depth. Below 2500 m, a steady

decline (~ 3o C/100 m) is noted in all the five temperature logs to the bottom of the well. The

increase in temperature from the surface to about 2250 m depth coincides with the trend of

increasing alteration grade. A prograde change in alteration grade, cutting across different

alteration zones, is observed to a depth of about 1798 m. This is consistent with the transition

from the low-temperature alteration minerals, zeolites, smectites and chalcedony to

relatively high-temperature minerals, such as quartz, albite, wairakite, illite, chlorite and

epidote (Figure 14).

Five small feed zones have been recognised from the study well. The classification criterion

is based on the relative size of kink in temperature recovery logs and the extent of geological

observations. These are located below the production casing at depths between 1260-1312

m, 1840-1900 m, 2070-2162 m, 2210-2276 m and 2360-2444 m. The feed zones are clearly

evident from the kinks in temperature logs (Figure 21), and are supported by both drilling

and geological observations. Feed zones noted above the production section (e.g. between

96-190 m) are of no consequence since this region has been completely cased off. However,

such feed zones are costly to the drilling operations because they require large tonnes of

cement to seal. Therefore, it is necessary to map and note their geological extent as a

precaution to subsequent wells planned to be drilled on the same well pad. The feed zone

between 1260-1312 m is hosted within a highly vesicular and altered tuff, marked by

extensive pyritization. A remarkably high rate of penetration (4.8-10.2 m/hr) is observed

within this zone, lending support to the geological observations.

Figure 20: Downhole pressure recovery logs for well

OW-922.

44

Feed zone between 1840-1900 m, is hosted within a highly altered and extensive trachytic

formation (Figure 14), characterised by numerous loss zones. Also, a deflection both in the

temperature profile and ROP profile is noted within this zone. Feed zones between 2070-

2162 m and 2210-2276 m are located within highly vesicular and moderately to highly

altered trachyte, characterised by several minor (2-10 m) loss zones. The final feed zone

between 2360-2444 m is located at the lithological interface of trachydacite and rhyolite. It

is also characterised by moderately to highly altered rhyolite, intense calcite deposition and

several loss zones, ranging in size between 2 and 14 m. Temperature recovery profile

indicates the highest formation temperature (127oC) measured in the well, occur within this

interval, despite the zone being associated with scarcity of high temperature alteration

minerals. The observed feed zones are summarized in Table 4.

45

Figure 21: Feed zones in well OW-922 as deduced from temperature recovery logs,

drilling parameters and geological observations.

46

Table 4: Interpreted permeable zones based on geological observations.

Depth (m) Evidence based on geological observations Size

1260-1312 Highly altered and vesicular tuff

Prevalence of pyrite deposition

Small

1840-1900 Highly altered trachyte

Numerous circulation losses

Small

2070-2162 Highly altered and vesicular trachyte

Numerous circulation losses

Small

2210-2276 Highly altered and vesicular trachyte

Numerous circulation losses

Small

2360-2444 Lithological interfaces between trachydacite

and rhyolite.

Moderate to high alteration intensity

Several circulation losses

Intense calcite deposition

Small

4.4 Alteration and measured formation

temperatures

Studies of temperature dependent alteration minerals, temperature logging (direct down-hole

measurements) and fluid inclusion homogenization temperatures (Th) have successfully

been applied to assess the physico-chemical conditions and temporal evolution of

geothermal reservoirs in active geothermal fields (e.g. Marks et al., 2010). A hydrothermal

alteration sequence provides significant insights into the history of a geothermal system.

There is always some uncertainty whether the observed alteration mineral assemblages

reflect the present formation temperatures or are related to some past thermal events. Like

in many geothermal systems, hydrothermal alteration minerals in the Olkaria geothermal

field have successfully served as important indicators of sub-surface thermal changes (e.g.

Lagat, 2004; Mwangi, 2012; Okoo, 2013).

A correlation of the three methods leads to reconstruction of a more plausible and realistic

model of a geothermal system. Fluid inclusion measurements were, however, not conducted

on samples from the study well. As a result, reconstruction of the thermal history of the

geothermal system around well OW-922 is based exclusively on inferred alteration and

measured formation temperatures (Figure 22). Alteration temperature is used to present

temperature in a geothermal system, and in many cases is assumed to represent the maximum

long-term temperature state. In contrast, measured formation temperature represents the

current state of the system. Alteration mineral curve plotted in Figure 22 relied on the first

appearance of alteration minerals observed, namely quartz, wairakite and epidote. The three

minerals have their first appearence at depths of 476 m, 494 m and 1542 m, respectively.

Temperature profile was plotted based on a 49 day heat-up period, i.e. the last temperature

profile measured in the well.

47

4.5 Whole-rock chemistry

Bulk-rock chemical analysis is important in understanding the chemical diversity of the

samples from the study well and the conditions under which the magma was generated. To

achieve this, plots have been made, including TAS diagram, Al2O3 vs. FeO and several

Harkers' variation diagrams based on the chemical analytical results. Generally, any sample

has two types of elements: major and trace (including rare earth elements).

4.6 Major and trace element Compositions

Thirty representative samples were analysed for whole-rock chemistry (major and trace

elements) in the study well. Concentrations of 10 major element oxides and 13 trace elements

were determined using the ICP-OES procedure outlined in Appendix C.

Figure 22: Comparison of inferred alteration and measured

formation temperatures along with boiling point curve for well OW-

922.

48

4.6.1 Major elements

Rocks in well OW-922 exhibit significant variations in major element compositions. Mafic

and intermediate rock suites have striking similarities, depicting a homogeneous trend for

FeO, MgO, and CaO, all showing a continuous decrease with increase in SiO2 content for

most of the samples (Figure 23). FeO shows a limited range (10.2-10.6 wt.%) in the mafic

and intermediate rocks and, slightly low abundance (3-8 wt.%) in the highly evolved rocks.

MgO is highly depleted in the highly evolved rocks (0.03-0.7 wt.%), whereas, the content

ranges between 3-6 wt.% in the mafic and intermediate rocks. CaO has a more or less similar

pattern to FeO, depicting a restricted range of between 6-11 wt.% in the mafic and

intermediate rocks and declining significantly to an average of about 1 wt.% in the most

silicic rocks.

TiO2 exhibits a limited range in the mafics and intermediate rocks (1.83-2.08 wt.%) and

decreases rapidly with increasing SiO2 content (Table 5). P2O5 displays considerable scatter

and no obvious variation trend, even though its contents are observed to decrease with

increase in SiO2 concentration. The scatter is more pronounced in the basic and intermediate

fields. Like TiO2, P2O5 is marked by a narrow range in the mafic and intermediate rocks

(0.18-0.60 wt.%). The contents decrease from about 0.60 wt.% in the mafic rocks to just

under 0.04 wt.% in the more silicic rocks, that is at approximately 69-73 wt.% SiO2. On the

other hand, MnO shows a nearly flat trend without any systematic variation with SiO2

(Figure 23). The elements' contents are fairly constant between 0.13 and 0.28 wt.%. As a

result, little emphasis is given on this element in interpretation of the data.

Al2O3 contents range from 10.3-17.04 wt.% in all the rock types. The element initially

exhibits a more or less similar picture as MnO, but displays modest decline with increasing

SiO2 contents for most of the samples. Na2O and K2O correlate positively with SiO2 at the

initial stage. The two oxides have a very similar pattern even though sodium increases

relatively more than potassium. The concentration of the two oxides increases and attains a

peak at approximately 68 and 64 wt.% SiO2 for potassium and sodium, respectively,

followed by a slight decrease with increasing SiO2 contents. The decrease is more prominent

in sodium, particularly for the samples at the greatest depths (i.e. between 2420 and 2990 m)

in the well. In contrast, K2O content appears to flatten out with a further increase in SiO2

concentration. The two elements are generally lower in the mafics and intermediate rocks

compared to the evolved rocks. In particular, Na2O concentration is distinctly higher than

K2O, with the exception of a few samples in the trachytes and trachydacites. The composition

of the two elements in rhyolites is, however, quite variable. Peralkalinity index (i.e. P.I. =

mol. [Na+K]/Al) of the samples analysed ranges between 0.26 and 1.03.

49

1

2

3

4

5

6

7

8

45 50 55 60 65 70 75

Na 2

O (

wt.

%)

6

8

10

12

14

16

18

Al 2

O3

(wt.

%)

0

2

4

6

8

45 50 55 60 65 70 75

K2O

(w

t.%

)

0,0

0,1

0,2

0,3

45 50 55 60 65 70 75

Mn

O (

wt.

%)

0,0

0,5

1,0

1,5

2,0

2,5

45 50 55 60 65 70 75

TiO

2(w

t.%

)

0

2

4

6

8

10

12

14

45 50 55 60 65 70 75

CaO

(w

t.%

)

0

2

4

6

8

45 50 55 60 65 70 75

MgO

(w

t.%

)

SiO2 (wt.%)

0

2

4

6

8

10

12

45 50 55 60 65 70 75

FeO

(w

t.%

)

SiO2 (wt.%)

Figure 23: Harker diagrams showing major element variations against SiO2 for samples

from well OW-922.

50

Figure 23 (continued): Harker diagrams showing major element oxides and CaO/ALK

variations agains SiO2 for samples from well OW-922.

4.6.2 Trace elements

The term trace element lacks a rigid definition, even though several authors (e.g.

Harðarson, 1993) have used this term in many instances to refer to those elements present

in rocks in concentrations of a few parts per million (ppm). The significant features of the

trace element concentrations observed from the variation diagrams (Figure 24) as

differentiation progresses from the mafic to silica rich rocks are: (1) the progressive

depletion in the concentrations of the transition metals, Sc, Co, V, Cr, Ni, and, Sr with

increase in SiO2 contents, (2) Incompatible trace elements (ITEs), such as the HFSE, Y and

Zr; Rb and LREE, La, show significant enrichment with increase in SiO2 contents.

Vanadium decreases approximately from 300 ppm to 168 ppm from the mafic to

intermediate rocks. Subsequently, an abrupt drop in V content down to 0.5 ppm is observed

in the highly evolved rocks. Cr content is high in the mafic rocks (132 ppm) but exhibit a

remarkable decline in the intermediate rocks (17-59 ppm). The concentration further

decreases to between 2-27 ppm in the silicic rocks. Sc vs. SiO2 plot shows a small but

consistent decrease in Sc content between the mafic and intermediate suites (47-25 ppm)

until about 69 wt.% SiO2 where the drop is rather steep (0.2-0.3 ppm). The rhyolitic rocks

are more depleted in Sc compared to the other evolved rock suites. A fair amount of scatter

is apparent in Sr concentration in the mafic and intermediate rocks as well as in two

trachytic samples (1878 m and 1922 m). The element shows a decrease from 376 ppm to

256 ppm as SiO2 increases to about 55 wt.%. A significant decrease to about 8 ppm,

thereafter follows in the more evolved derivatives. Two trachytic samples at 1878 m and

1922 m depth, however, contain anomalous Sr contents, 307 ppm and 344 ppm at relatively

high SiO2 contents, i.e. 58 wt.% and 63 wt.%, respectively.

Rubidium shows high abundance (123-226 ppm) with increase in SiO2 (64-73 wt.%). A

significant decline (≤ 78 ppm) is observed in the mafic and intermediate rocks,

characterised by SiO2 range of between 49-55 wt.%. Y displays a very similar trend to Rb,

showing a gradual increase with increasing SiO2 concentration. In particular, it is more

enriched (101-304 ppm) between 67-73 wt.% SiO2. La follows similar pattern as Rb and

Y. The element shows a concentration of between 110-424 ppm at SiO2 content of between

63-73 wt.%. The contents, however, fall to < 80 ppm in the mafic and intermediate rocks.

0,0

0,2

0,4

0,6

0,8

45 50 55 60 65 70 75

P2O

5(w

t.%

)

SiO2 (wt.%)

0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

45 50 55 60 65 70 75

CaO

/ALK

SiO2 (wt.%)

51

Zr shows significant increase with increasing SiO2 content but scatters mostly between

2793 and 2995 ppm. Ba, on the other hand exhibits wide scattering in whole-rock

compositions with no well defined trend. The element occurs in considerable abundances

in the intermediate rocks, even though exceptionally high abundances (1775-2238 ppm)

are observed in the trachytes. However, Ba is strongly depleted in the rhyolites, in most

cases falling below 20 ppm. Results of major and trace element concentrations are

presented in Table 5.

0

100

200

300

400

45 50 55 60 65 70 75

Sr (

pp

m)

0

10

20

30

40

50

45 50 55 60 65 70 75

Sc (

pp

m)

0

100

200

300

400

45 50 55 60 65 70 75

V (

pp

m)

0

40

80

120

160

45 50 55 60 65 70 75

Cr

(pp

m)

0

20

40

60

80

100

120

45 50 55 60 65 70 75

Ni (

pp

m)

SiO2 (wt.%)

0

10

20

30

40

50

60

45 50 55 60 65 70 75

Co

(p

pm

)

SiO2 (wt.%)

Figure 24: Variation diagrams for a range of trace elements plotted against the

index of differentiation, SiO2 for samples from well OW-922.

52

4.7 Classification of rock types

Petrologists have devised several schemes for the chemical classification of non-pyroclastic

volcanic rocks. In spite of that, none has been found fully satisfactory. Many systems were

in use before classification based on the amount of alkali (Na2O + K2O) versus silica (SiO2)

diagram in igneous rocks, popularly known as the “TAS diagram” (Figure 25), was

eventually accepted by the International Union of Geological Sciences (IUGS). The

variation of silica and total alkalis is a highly useful discrimination technique. The total alkali

0

100

200

300

400

45 50 55 60 65 70 75

Y (p

pm

)

0

100

200

300

400

500

45 50 55 60 65 70 75

La (

pp

m)

0

500

1000

1500

2000

2500

45 50 55 60 65 70 75

Ba

(pp

m)

0

50

100

150

200

250

45 50 55 60 65 70 75

Rb

(p

pm

)

0

500

1000

1500

2000

2500

3000

3500

45 50 55 60 65 70 75

Zr (

pp

m)

SiO2 (wt.%)

0

40

80

120

160

200

45 50 55 60 65 70 75

Cu

(p

pm

)

SiO2 (wt.%)

Figure 24 (continued) : Variation diagrams for a range of trace elements plotted against

the index of differentiation, SiO2 for samples from well OW-922.

53

axis essentially distinguishes alkaline from sub-alkaline rocks types, whereas the silica axis

distinguishes between evolved and primitive rock types. An important contributing factor

that led to the acceptance of this method was its simplicity, which is an essential aspect of a

classification scheme for volcanic rocks (Le Bas et al., 1986).

Generally, the compositional variation seen on a TAS diagram may be regarded as a result

of two basic processes, namely partial melting and fractional crystallisation, even though

magma mixing can be important in some cases. In this respect, decreasing degree of partial

melting results in increased total alkalis and lower SiO2. Fractional crystallisation, on the

other hand brings the rocks to both higher alkali and silica content (e.g. Harðarson, 1993).

In this section, petrochemical results from the study well are presented and compared with

the results of the previously studied wells (Musonye, 2015) aimed at understanding the

chemical composition and petrogenetic mechanisms responsible for the evolution of OVC

sub-surface rocks. Additionally, the results are compared with the findings for surface

samples for Olkaria and the minor eruptive centres surrounding the complex.

4.7.1 TAS classification scheme

In the TAS classification scheme, the concentrations of the major elements have been re-

calculated to 100% on a volatile-free basis to ensure comparability and are plotted as such

in the variation diagrams. From the classification of the rocks in the TAS system shown in

Figure 25, it is apparent that rocks from well OW-922 show a range of compositions. The

sub-alkaline rocks vary from basalt through trachydacite to rhyolite, whereas, the alkaline

rocks range from basaltic trachyandesite through tracyandesite to trachyte.

Figure 25: TAS diagram showing the compositional range for sub-surface rocks from

well OW-922, according to Le Bas et al. (1986).

54

The prevalence (~ 85-90%) of evolved rocks is noticeable, depicting a relatively wide range

in SiO2 (59-73 wt.%). Basic and intermediate rocks constitute about 10% of the intersected

succession, and have a SiO2 range of between 49.7-55 wt.% (Table 5). A large volume of

the salic differentiates cluster to a great extent in the trachydacite, rhyolite and trachyte

fields, with the latter being the dominant rock type. A few samples are, however, transitional

in composition, falling along the boundary separating the alkaline from sub-alkaline series.

Based on petrographic analysis, the rocks from well OW-922 are variably porphyritic with

plagioclase and clinopyroxene as the dominant phenocrysts phases, together with olivine and

opaques (Fe-Ti oxides) in the mafic to intermediate samples. In the more salic varieties, K-

feldspar (sanidine) is the predominant phenocryst phase, although plagioclase and

clinopyroxene (aegerine-augite) are also noted in considerable amounts.

4.7.2 Al2O3 vs. FeO classification scheme

This is a classification scheme put forward by Macdonald (1974), using the correlation

between alumina and total iron (as FeO) plot to evaluate the relationship of the samples

based on the degree of peralkalinity. Peralkaline rocks are basically divided into comendites,

pantellerites, comenditic trachytes and pantelleritic trachytes (Macdonald and Scaillet,

2006). By definition, comenditic rocks refers to the mildly peralkaline varieties, whereas,

the pantelleritic counterparts, refers to more strongly peralkaline and more Fe-rich rocks

(FeO>Al2O3) (Macdonald, 1974). None of the analytical samples fall under the definition of

pantelleritic, as in all cases, Al2O3 is observed to be greater than FeO (Table 5).

Based on this classification scheme (Figure 26), it is evident that rocks from the study well

traverse the division between comenditic trachyte, pantelleritic trachyte and pantellerite,

with the last one being the dominant variety. No samples plot within the comendite field.

Most of the trachytes plot in the comenditic trachyte field, with a few exceptions plotting as

pantelleritic trachyte and pantellerites. Rhyolites, on the other hand plot within the

pantellerites field. From the Al2O3 vs. FeO plot, it is clear that the differentiation trends for

these samples are appreciably variable.

55

Figure 26: Al2O3 vs. FeO classification (modified from Macdonald, 1974) of sub-

surface rocks from well OW-922.

56

Table 5: Major and trace element results for well OW-922 based on ICP-OES analysis.

Major elements (wt.%)

Depth 864 m 966 m 1106 m 1142 m 1208 m 1332 m 1672 m 1704 m 1734 m 1808 m 1878 m 1922 m 1966 m 2056 m 2088 m

SiO2 61.85 63.57 69.17 69.36 49.67 62.74 54.62 65.85 64.65 64.89 63.59 58.99 64.71 65.26 65.55

Al2O3 14.50 13.84 10.50 10.30 14.94 15.34 15.10 13.70 15.30 14.40 17.04 14.76 14.02 12.96 13.32

FeO 8.08 8.29 8.11 8.05 10.66 7.06 10.21 7.65 6.34 6.26 3.31 8.88 7.32 6.87 7.18

MnO 0.27 0.28 0.24 0.24 0.21 0.25 0.21 0.23 0.21 0.20 0.13 0.21 0.27 0.24 0.25

MgO 0.25 0.03 0.05 0.06 6.84 0.35 3.38 0.42 0.37 0.33 0.56 0.68 0.51 0.59 0.22

CaO 1.96 1.31 0.57 0.53 11.67 1.17 6.64 1.44 1.10 1.14 2.02 2.58 1.02 1.47 0.54

Na2O 6.83 6.95 5.90 6.00 2.98 6.73 4.44 4.14 5.69 6.25 7.04 6.35 5.13 6.23 5.81

K2O 5.28 4.93 4.61 4.57 0.87 5.38 2.47 5.47 5.14 4.97 4.93 4.85 5.88 5.17 5.95

TiO2 0.75 0.64 0.42 0.44 1.83 0.62 2.08 0.82 0.93 1.04 0.91 1.67 0.88 0.93 0.92

P2O5 0.10 0.02 0.01 0.01 0.18 0.08 0.60 0.08 0.13 0.25 0.20 0.72 0.09 0.10 0.11

Total 99.87 99.87 99.58 99.56 99.85 99.72 99.76 99.79 99.85 99.72 99.73 99.68 99.82 99.82 99.85

Trace elements (ppm)

Ba 482.52 26.48 0.00 0.00 255.88 24.22 1094.92 41.20 55.68 1775.59 1777.61 2238.54 42.70 45.74 131.01

Co 5.05 3.15 1.45 1.72 52.93 4.63 38.92 5.91 6.59 7.30 7.39 16.15 5.91 7.23 6.38

Cr 4.77 3.85 15.75 26.69 131.78 2.33 17.39 6.16 2.72 5.41 5.77 6.20 4.87 4.21 3.80

Cu 14.41 16.34 20.83 22.69 110.58 25.71 35.54 24.57 16.78 16.81 13.50 14.04 18.61 20.50 16.80

La 82.72 139.07 424.45 422.93 19.76 317.50 75.51 231.23 158.38 70.42 65.98 48.92 146.52 159.60 110.13

Ni 0.00 0.00 64.13 91.44 54.37 0.00 1.38 0.00 0.00 5.32 1.35 4.81 1.92 1.85 1.00

Rb 97.11 93.39 150.54 132.69 20.08 118.30 78.42 139.10 98.73 152.38 73.15 100.18 170.05 128.52 169.74

Sc 3.75 1.09 0.20 0.30 47.57 1.99 25.40 3.43 4.98 17.24 3.41 16.81 4.73 4.87 5.67

Sr 24.23 8.66 13.94 18.36 303.93 15.71 376.84 18.81 14.28 42.69 344.45 307.00 23.05 24.74 17.10

V 0.56 2.79 19.56 20.94 300.20 14.13 168.63 9.16 5.71 3.85 14.98 5.58 5.64 6.15 1.93

Y 51.47 92.92 302.93 282.72 26.18 193.71 49.68 145.25 100.40 71.72 34.64 55.36 133.06 124.64 101.08

57

Zn 136.74 179.65 358.59 368.48 97.19 222.95 126.65 224.30 165.60 126.90 65.26 134.52 193.55 199.89 175.22

Zr 368.19 706.84 2793.51 2984.86 100.40 1895.18 349.16 1230.88 863.51 541.83 340.04 283.14 1080.77 1102.87 773.63

Total 1271.53 1274.23 4165.89 4373.81 1520.84 2836.36 2438.43 2080.02 1493.34 2837.47 2747.52 3231.27 1831.38 1830.80 1513.49

Table 5 (continued): Major and trace element results for well OW-922 based on ICP-OES analysis.

Major elements (wt.%)

Depth 2190 m 2234 m 2288 m 2300 m 2348 m 2360 m 2420 m 2482 m 2522 m 2570 m 2600 m 2668 m 2782 m 2802 m 2982 m

SiO2 66.32 64.66 67.21 67.99 67.02 68.11 73.26 71.19 70.38 69.71 67.80 67.79 55.14 65.98 64.33

Al2O3 14.10 13.81 12.72 12.39 14.43 12.20 10.92 11.59 12.21 11.44 13.11 13.24 11.92 14.67 14.99

FeO 6.34 6.92 7.32 6.50 6.58 8.37 5.97 7.15 6.74 6.97 7.05 6.82 10.43 5.96 6.54

MnO 0.24 0.26 0.27 0.22 0.20 0.27 0.19 0.23 0.24 0.25 0.27 0.25 0.25 0.23 0.23

MgO 0.47 0.52 0.38 0.35 0.47 0.39 0.13 0.17 0.22 0.28 0.21 0.32 5.03 0.33 0.69

CaO 1.36 0.98 0.74 0.78 2.06 0.99 1.02 0.71 0.66 0.99 0.88 0.63 6.73 0.99 1.58

Na2O 6.31 5.53 4.05 4.72 4.06 5.61 1.47 3.25 3.66 2.94 4.53 4.33 4.59 5.31 5.27

K2O 3.50 5.97 6.24 5.98 3.78 3.13 6.12 4.83 4.93 6.51 5.20 5.50 3.81 5.28 4.98

TiO2 1.06 1.05 0.78 0.79 1.09 0.63 0.62 0.62 0.69 0.65 0.70 0.84 1.62 0.95 1.03

P2O5 0.14 0.15 0.08 0.06 0.11 0.04 0.04 0.01 0.04 0.03 0.05 0.07 0.19 0.13 0.15

Total 99.84 99.84 99.78 99.78 99.79 99.75 99.73 99.74 99.76 99.77 99.80 99.79 99.71 99.82 99.79

Trace elements (ppm)

Ba 149.58 127.92 64.01 58.13 325.68 100.83 20.24 12.19 10.05 18.94 13.73 69.62 214.55 229.35 299.50

Co 8.01 7.94 5.05 4.97 12.48 4.78 2.58 3.22 3.64 3.36 3.93 5.43 38.67 7.32 10.41

Cr 4.16 4.99 3.45 12.34 20.22 5.93 3.90 3.37 3.52 2.88 5.28 4.05 59.80 4.49 5.08

Cu 17.22 26.86 22.41 15.38 19.64 25.67 16.05 15.08 15.86 21.31 15.04 16.59 161.25 30.68 25.51

La 121.69 110.69 162.27 172.77 162.11 226.33 261.40 260.69 228.78 198.92 185.74 169.70 181.88 144.24 155.43

Ni 2.00 5.38 1.15 7.55 9.71 2.54 2.35 1.10 3.92 2.90 2.74 2.11 112.51 34.59 11.70

Rb 134.79 173.93 182.65 189.47 147.49 176.99 183.63 171.98 226.73 135.79 133.92 123.74 76.88 144.93 173.04

58

Sc 5.71 6.28 4.30 3.92 4.86 1.58 2.34 2.38 3.27 3.38 3.95 3.92 25.14 5.53 6.23

Sr 40.72 21.41 22.16 20.88 77.13 32.11 20.81 14.97 14.72 18.34 13.76 22.37 256.51 23.65 76.20

V 5.10 4.23 8.24 7.46 31.38 14.51 10.84 10.79 7.96 8.21 7.25 9.06 200.34 5.18 21.44

Y 100.80 99.82 140.46 137.88 124.71 178.46 176.06 185.15 151.88 139.13 131.22 141.02 125.19 95.25 106.83

Zn 148.91 175.62 230.49 199.54 201.56 283.65 210.56 227.88 214.70 245.66 215.22 203.28 248.31 162.09 153.31

Zr 837.44 810.00 1355.26 1357.33 943.30 1487.53 1807.54 1733.52 1479.58 1479.69 1275.09 1365.92 1189.73 934.39 1027.02

Total 1576.14 1575.07 2201.89 2187.62 2080.27 2540.91 2718.29 2642.31 2364.62 2278.52 2006.88 2136.81 2890.78 1821.69 2071.70

59

4.8 Comparison of well OW-922 with surface

and other sub-surface samples

This section presents a comparison of the geochemical results from well OW-922 with

surface samples from Olkaria and other minor eruption centres located along the periphery

of the complex (Figure 8). Surface data from Olkaria, together with the minor eruption

centres were acquired from previous studies of Macdonald et al. (2008), Marshall et al.

(2009) and Clarke et al. (1990). Majority of the surface samples are considered quite fresh

and apparently devoid of alteration. Additional sub-surface whole-rock chemistry data have

been obtained from Olkaria, published by Musonye (2015).

4.8.1 Comparison with surface samples

From Figure 27, it is immediately evident that samples from wells OW-922, 905A, 910 and

917 plot in more or less similar fields to the surface samples from Olkaria and neighboring

minor eruption centres. The Lolonito, Ndabibi and Akira basalts (Figure 7) plot in a similar

field to basalt and basaltic trachyandesite from the study well and the other three wells.

Equivalently, the Gorge Farm and Broad Acres rhyolites, Olkaria comendites (Figure 8) and

Tandamara trachytes (Figure 7), plot mostly in a similar position as the trachyte and rhyolites

from well OW-922.

LB NB AB

OC TT GF BA

OC- Olkaria Comendite TT- Tandamara Trachyte GF- Gorge Farm BA- Broad Acres LB- Lolonito Basalt NB- Ndabibi Basalt AB- Akira Basalt

Figure 27: TAS classification scheme showing comparison of sub-surface samples from

wells OW-922, 905A, 910 and 917 and, surface samples from Olkaria and neighboring

minor eruption centres (represented by broken lines).

60

4.8.2 Comparison with other sub-surface samples

Using selected major element oxides and trace element variation diagrams (Figures 28 and

29), a comparison has been made to check for any connection in the evolution mechanism(s)

between wells 905A, 917, 910 and 922. Once again, SiO2 has been used as the index of

differentiation. It is noticeable that the ferromagnesian major element oxides (FeO, MgO,

CaO) as well as TiO2 and P2O5 in the other three wells behave in a manner akin to well OW-

922 (Figure 23), showing a gradual decline with increasing SiO2 values. For K2O and Na2O,

the trend starts with a positive slope and bends to a negative one at ~ 65 wt.% and ~ 62 wt.%

SiO2 respectively. The change in slope is far more pronounced in Na2O than K2O. Similarly,

V, Sr, Sc and Co are anti-correlated with SiO2 (Figure 29), and show a very similar overall

picture to the ferromagnesian major oxides, pointing to their compatible behavior. On the

other hand La, Y, and Zr exhibit positive correlations with SiO2. It was observed that rocks

from wells OW-905A, 910 and 917 are slightly more enriched in SiO2 (44-80 wt.%) than

corresponding rocks from the study well (49-73 wt.%). In summary, comparison of the

geochemical results for the selected major and trace elements for the four wells demonstrate

a perfect overlap across the whole range of compositions.

0

4

8

12

16

40 50 60 70 80 90

FeO

(w

t.%

)

SiO2 (wt.%)

OW-922 OW-905A OW-910 OW-917

0

2

4

6

8

40 50 60 70 80 90

MgO

(w

t.%

)

SiO2 (wt.%)

OW-922 OW-905A OW-910 OW-917

0

4

8

12

16

40 50 60 70 80 90

CaO

(w

t.%

)

0

1

2

3

4

5

40 50 60 70 80 90

TiO

2(w

t.%

)

Figure 28: A comparison for variation in concentration of selected major element oxides

for wells OW-922, 905A, 910 and 917.

61

0

2

4

6

8

10

40 50 60 70 80 90

Na 2

O (

wt.

%)

SiO2 (wt.%)

OW-922 OW-905A OW-910 OW-917

0

2

4

6

8

40 50 60 70 80 90

K2O

(w

t.%

)

SiO2 (wt.%)

OW-922 OW-905A OW-910 OW-917

0

200

400

600

800

1000

40 50 60 70 80 90

Sr (

pp

m)

0

100

200

300

400

40 50 60 70 80 90

V (

pp

m)

0

10

20

30

40

50

40 50 60 70 80 90

Sc (

pp

m)

SiO2 (wt.%)

OW-922 OW-905A OW-910 OW-917

0

20

40

60

80

40 50 60 70 80 90

Co

(p

pm

)

SiO2 (wt.%)

OW-922 OW-905A OW-910 OW-917

Figure 29: A comparison for variation in concentration of selected trace elements for

wells OW-922, 905A, 910 and 917.

Figure 28 (continued): A comparison for variation in concentration of selected major

element oxides for wells OW-922, 905A, 910 and 917.

62

4.9 Microprobe analysis

Selected samples of epidote and chlorite from wells OW-922, 905A, 910 and 917 were

analysed with an electron microprobe (EMP). An important aspect of this analysis is to

assess the chemical composition and the compositional range of epidote and chlorite in the

mentioned wells. Interpretations made have been correlated with compositional ranges

observed in other geothermal systems (e.g. Bird et al., 1988; Marks et al., 2010; Lonker et

al., 1993; Bragason, 2012; Bird and Spieler, 2004). The first appearance of epidote and

chlorite has commonly been used as a criterion for setting production casing in production

wells in the Olkaria geothermal field. The selected epidote and chlorite samples analysed in

the present study occur either as open space fillings in vesicles and fractures or as a

replacement of original rock forming minerals, mostly plagioclase and sanidine. Associated

alteration minerals were separated and not simultaneously analysed alongside these two

minerals.

The mineral analyses were performed with a JEOL JXA-8230 electron microprobe located

at the Institute of Earth Sciences, University of Iceland using carbon coated finely polished

thin sections. The elements analysed were Si, Ti, Al, Fe, Mn, Mg, Ca, Na, K and Sr.

Quantitative analyses with wavelength-dispersive spectrometers (WDS) were performed at

an accelerating voltage of 15.0 keV, a cup current of 5.0 nA, and a beam diameter of between

5 to 10 µm. Counting time on peak was 40 seconds for all elements, except Na (30 seconds).

Background counting time varied for different elements. In the case of chlorite Si, Al, Fe,

Mn and Mg backgrounds were counted for 20 seconds, those of Ti, Ca, Sr and K for 40

seconds and Na 15 seconds. Background counting times for epidote analyses were 20

seconds for Si, Al, Fe, Mn and Ca, 40 seconds for Ti, Mg, K and Sr, and 15 seconds for Na.

Standards used were natural mineral standards. Structural formulae were calculated on the

basis of 24 oxygens. It should be noted that Fe is calculated as ferrous and ferric iron in

chlorite and epidote, respectively, and in all cases, Mn is calculated as divalent. Additional

data sets for this study are shown in Appendix D.

0

500

1000

1500

2000

2500

3000

3500

40 50 60 70 80 90

Zr (

pp

m)

SiO2 (wt.%)

OW-922 OW-905A OW-910 OW-917

0

100

200

300

400

500

40 50 60 70 80 90

Y (p

pm

)

SiO2 (wt.%)

OW-922 OW-905A OW-910 OW-917

Figure 29 (continued): A comparison for variation in concentration of selected trace

elements for wells OW-922, 905A, 910 and 917.

63

Epidote [Ca2FexAl3-xSi3O12(OH)] is an important index mineral in geothermal systems. The

first appearance of this mineral demonstrates convincingly temperatures equal to or in excess

of 240oC. Detailed analysis of epidote compositions has been carried out in a number of

geothermal systems, e.g. Salton Sea, Cerro Prieto, Los Azufres, Reykjanes, just to name a

few with high level of success. However, no quantitative analysis of epidote have been done

in the Olkaria geothermal system despite epidote being such an important mineral. The

ability to accurately determine element concentrations of epidote by EMP analysis is crucial

for a complete understanding of the chemical composition and changes in the compositional

range.

There are three octahedral sites, M(1), M(2) and M(3), in monoclinic epidote where

significant exchange of Fe3+ and Al3+ occurs. Ferric iron readily substitutes for Al3+ in both

the M(1) and M(3) sites. Several studies (e.g. Arnason et al., 1993) have indicated that Fe3+

is preferentially incorporated in the energetically larger and more distorted M(3) site and, to

a lesser degree, in the smaller and more regular M(1) site. At relatively low temperature, site

M(2) is nearly free of Fe3+ and mostly contains Al3+. It is only at high temperature that ferric

iron enters the M(2) site.

The compositional range of most natural epidotes in active geothermal systems is closely

described as a binary solid solution between the end-members clinozoisite [Ca2Al3Si3O12

(OH)] and pistachite [Ca2Fe3Si3O12(OH)]. Hence, the composition variation is dominated by

the substitution of Fe3+ and Al3+ in the octahedral sites (Marks et al., 2010). Epidote

composition is commonly expressed as the mole fraction of the pistachite component. The

notation Xps has been adopted to denote the mole fraction of pistachite [Xps = nFe / (nFe +

nAl)] where nFe and nAl designate cation proportions in each formula unit.

Epidote paragenesis is controlled by a variety of intensive and extensive thermodynamic

variables characterising hydrothermal systems. These include CO2 and O2 fugacities (ƒCO2,

ƒO2), coexisting mineral assemblages, changes in fluid chemistry, bulk rock composition

and temperature (e.g. Bird and Spieler, 2004). The functional relationships between epidote

solid solution compositions and changes in these variables depend on the stoichiometry of

reactions and enthalpies (Arnason et al., 1993).

Epidote was analysed in four different samples in this study; two samples derived from two

different depth intervals in well OW-910 and, one sample from a single depth range in each

of wells OW-905A and 922. No epidote sample was identified in well OW-917 for

microprobe analysis. It is important to note that, due to the paucity of epidote samples

analysed from wells OW-922 and 905A, it has not been possible to correlate the stratigraphy

in the three wells on the basis of major and minor element contents. Such comparison would

have been of great interest so as to enable critical assessment of the compositional range.

Therefore, the compositional range discussed in the succeeding sections is based largely on

analytical results from well OW-910. Epidote samples analysed were noted to occur in

association with chlorite, titanite, magnetite and albite (e.g. Figures 30 and 31). A

representative selection of chemical analyses illustrating the range in the composition of

epidote is shown in Table 6. As iron in epidote is dominantly in the trivalent state (Fe3+),

total Fe is reported as ferric iron (Fe2O3) as mentioned earlier. Low totals are at least partly

caused by the presence of the hydroxyl ion (OH−) in the epidote structure, which is not

detected by the microprobe.

64

Table 6: Representative EMP analyses of epidote from wells OW-922, 910 and 905A.

Well OW-922 OW-922 OW-910 OW-910 OW-910 OW-910 OW-905A OW-905A

Depth 2782 m 2782 m 864 m 864 m 1236 m 1236 m 1560 m 1560 m

Analysis as Oxides, wt.%

SiO2 37.48 37.50 37.32 38.14 36.16 35.77 37.12 36.39

TiO2 0.00 0.03 0.24 0.02 0.47 0.49 0.14 0.08

Al2O3 24.62 24.71 22.21 25.57 8.03 7.23 23.10 23.48

Fe2O3 12.71 12.48 16.14 11.86 19.91 21.11 14.86 13.78

MnO 0.11 0.17 0.14 0.04 0.48 0.39 1.53 1.05

MgO 0.11 0.05 0.09 0.06 0.35 0.29 0.05 0.02

CaO 22.97 22.81 23.21 23.15 32.59 33.05 21.50 22.16

Na2O 0.02 0.03 0.00 0.00 0.02 0.01 0.00 0.01

K2O 0.04 0.04 0.00 0.01 0.03 0.05 0.01 0.01

SrO 0.36 0.32 0.10 0.22 0.00 0.01 0.05 0.01

Total 98.43 98.13 99.46 99.08 98.04 98.41 98.35 97.00

Structural formulae based on 24 Oxygens

Si 5.694 5.707 5.681 5.721 5.961 5.916 5.693 5.647

Ti 0.000 0.003 0.028 0.002 0.058 0.061 0.016 0.010

Al 4.409 4.432 3.984 4.521 1.559 1.409 4.175 4.295

Fe 3+ 1.454 1.429 1.849 1.339 2.471 2.628 1.715 1.609

Mn 0.015 0.021 0.018 0.005 0.067 0.055 0.199 0.139

Mg 0.024 0.012 0.021 0.014 0.086 0.070 0.011 0.005

Ca 3.740 3.720 3.785 3.721 5.756 5.857 3.534 3.684

Na 0.006 0.010 0.000 0.001 0.005 0.004 0.000 0.002

K 0.008 0.007 0.001 0.002 0.006 0.010 0.001 0.002

Sr 0.032 0.028 0.009 0.020 0.000 0.001 0.004 0.001

Total 15.382 15.370 15.375 15.346 15.970 16.012 15.348 15.393

65

Figure 30: BSE image of epidote from 864 m in well OW-910. Compositional

heterogeneity is clearly seen in this figure with Al-rich sections appearing darker

and Fe-rich sections lighter in color. Numerous compositional zonation patterns

are represented by Al-rich rims and Fe-rich cores. Also, note the titanite inclusions

in epidote.

Figure 31: BSE image of epidote filling a vesicle from 1560 m in well OW-905A.

Epidote is observed occuring as aggregates of small columnar cystals. Quartz is

also noted growing in the vesicle. Black areas in the lower right hand corner, at

the centre and within quartz are epoxy.

66

The molar proportion of pistachite in epidote from the three wells ranges between 0.23-0.65,

as shown in the table below.

Table 7: Range and average molar proportion of pistachite (Xps) component in epidote.

Well Depth (m) Molar ratio of pistachite

nFe / (nFe + nAl)

range average

OW-922 2782 0.23 – 0.24 0.235

OW-910 864 0.23 – 0.32 0.275

OW-910 1236 0.61 – 0.65 0.630

OW-905A 1560 0.29 – 0.27 0.280

Chemical analyses presented in Table 6 indicate that the epidotes are relatively Al-rich types,

with the exception of epidote from 1236 m in well OW-910, which is Fe-rich. There is a

slight inhomogeneity in composition with respect to the Fe3+/Al3+ ratio of the epidotes

analysed. A narrow intragrain chemical compositional change exists for the major and minor

oxides for the three wells. However, intergrain chemical compositional change is more

pronounced in well OW-910, based on two samples derived from two different depth

intervals (864 m and 1236 m). Here, a systematic change in the composition of epidote with

increasing depth is clearly noticeable. This is illustrated by the change in the contents of the

two dominant elements, i.e. Al and Fe, that predominantly dictate the variation in the

composition of epidote (Figures 32 and 33). Al2O3 contents show a significant decline from

~ 24 to ~ 8 wt.% on average, whereas, Fe2O3 contents display a gradual increase from 14 to

~ 21 wt.% on average with depth. This is consistent with increasing average molar ratio of

pistachite component with depth (Table 7). Compositional zoning, caused by variable Fe3+-

Al3+ substitution and representing phase disequilibrium, is a common feature of epidotes

analysed in geothermal systems (Arnason and Bird, 1992). Other minor elements of

importance observed to be in variable concentration in epidote, though, to a lesser extent,

are Sr and Ti. These elements, like Al and Fe, are enriched on the rims and cores of epidote,

respectively, and are clearly visible on intragrain zonation.

67

Figure 32: A WDS elemental map showing typical compositional zoning of epidote from

864 m in well OW-910. Grayscale map (a) represent the original composition, whereas,

maps (b), (c) and (d) denote Fe, Sr and Al levels respectively.

68

Chlorites are basically hydrous silicates of aluminium, magnesium and/or iron, represented

by the structural formula (Mg,Al,Fe)12 (Si,Al)8 O20 (OH)16. Common cation substitution

involves replacement of Si4+ by Al3+ in a tetrahedral site, and, a limited substitution of R2+

by Al3+ in the octahedral site to preserve charge balance, where R2+ essentially is a

summation of Mg2+ + Fe2+ + Mn2+. Analyses of chlorite for this study were made from six

samples, two each derived from different depth intervals in wells OW-922 and 910. From

wells OW-905A and 917, one sample was analysed, representing a single depth range in

each case. Representative chemical analyses of chlorite are presented in Table 8. A review

of chlorite analyses (e.g. Dubois and Martinez-Serrano, 1998) shows that ferric iron typically

constitutes < 5% of total Fe. Therefore, it is appropriate to report Fe as divalent.

It should be noted that chlorite has a strong tendency to form very fine-grained mats of

needle-shaped mineral grains (e.g. Figure 34). This along with the relatively rough, uneven

surface of the chlorite mats results in relatively poor analyses. As chlorite is highly hydrated,

totals will always be low, but this can also mean that the mineral is relatively unstable under

the electron beam, even with a relatively low probe current and wide beam diameter. Hence,

the results of the EMP analyses of chlorite have to be interpreted with some caution. Most

Figure 33: A WDS elemental map showing Fe (b) and Al (d) distribution within a single

epidote grain from 2782 m in well OW-922. The darker and lighter sections in the original

compositional map (a) indicates Al and Fe-rich regions, respectively. Map (c) shows Sr

levels. Compositional zoning is not obvious as is the case in Figure 32.

69

published studies (e.g. Lonker and others, 1993) have shown that variations in the chemical

composition of chlorites are largely dependent on temperature and changes in the

composition of the host rocks. Changes in ƒO2 of geothermal fluids at depth is another

important factor which cannot be discounted, even though it plays a limited role. Variations

in host rock composition mainly affects distribution of Fe in chlorite phases, consequently

producing some changes in Fe-Mg occupancy (Bryndzia and Scott, 1987).

Table 8: Representative EMP analyses of chlorite for wells OW-922, 910, 917 and 905A.

Well OW-922 OW-922 OW-910 OW-910 OW-917 OW-905A

Depth 2236 m 2782 m 864 m 1236 m 1870 m 1560 m

Analysis as Oxides, wt.%

SiO2 22.20 22.12 25.02 30.45 19.58 25.61

TiO2 0.10 0.06 0.03 0.11 0.04 0.00

Al2O3 14.86 13.85 12.95 16.90 16.33 19.16

FeO 30.50 32.97 12.60 17.62 34.55 34.41

MnO 0.12 0.33 0.19 0.28 0.70 2.02

MgO 5.38 5.38 17.58 15.59 5.30 7.87

CaO 0.10 0.16 0.31 0.44 0.07 0.11

Na2O 0.75 0.18 0.03 0.09 0.47 0.02

K2O 0.32 0.13 0.02 0.01 0.58 0.02

SrO 0.03 0.00 0.00 0.02 0.00 0.02

Total 74.35 75.18 68.73 81.50 77.62 89.26

Structural formulae based on 24 Oxygens

Si 5.050 5.150 5.446 5.634 4.417 4.821

Ti 0.017 0.011 0.006 0.015 0.007 0.000

Al 3.985 3.727 3.323 3.686 4.342 4.250

Fe2+ 5.803 6.296 2.294 2.726 6.518 5.418

Mn 0.023 0.064 0.035 0.043 0.133 0.322

Mg 1.824 1.831 5.703 4.298 1.783 2.209

Ca 0.024 0.039 0.073 0.087 0.017 0.023

Na 0.329 0.080 0.013 0.031 0.207 0.009

K 0.092 0.039 0.005 0.004 0.167 0.005

Sr 0.004 0.000 0.000 0.002 0.000 0.003

Total 17.151 17.236 16.898 16.526 17.591 17.060

70

Chemical analyses of chlorites from the four wells indicate that they are Fe-rich varieties

except from 864 m depth in well OW-910 (Table 8). Most of the chlorites fall in the

brunsvigite compositional field going by the classification scheme proposed by Dubois and

Martinez-Serrano (1998). Si, Al, Fe and Mg are in appreciable amounts in chlorite in the

four wells, whereas Ti, Mn, Ca, Na, K and Sr are present as minor or trace elements.

Variations in Fe2+ / (Fe2+ + Mg) ratio in chlorite from hydrothermally altered rocks (Table

9) has in many cases (e.g. Lonker and others, 1993) been applied to infer permeable zones

in a well.

Table 9: Molar ratio of chlorite from wells OW-922 and 910.

Well Depth (m) Molar ratio of chlorite

nFe / (nFe + nMg)

OW-922 2236 0.76

OW-922 2782 0.77

OW-910 864 0.29

OW-910 1236 0.39

Figure 34: BSE image of chlorite replacing plagioclase at 1870

m in well OW-917.

71

5 Discussion

5.1 Lithological units

Well OW-922 is a step-out vertical well drilled to a total depth of 2990 m CT to the east, in

the plain between Olkaria Domes field and Longonot volcanic centre. The main purpose of

drilling the well was to appraise the geothermal resource potential in the area, so as to expand

production drilling outside the already congested proposed Olkaria caldera structure. This,

also comes at a time of renewed focus by the Kenyan government on accelerating

development and utilisation of geothermal resources for electricity generation in the country.

Based on analysis of drilling cuttings through binocular and petrographic microscopy and

ICP-OES, the sub-surface strata penetrated by the investigation well comprise pyroclastics,

tuffs, rhyolite, trachyandesite, basaltic trachyandesite, basalt, trachydacite and trachyte.

From the existing geologic well logs (KenGen in-house data), it is evident that no other well

drilled in Olkaria has intersected such thick sequence of pyroclastic deposits, extending

continuously for about 300 m, as is the case with the study well. The observation is perhaps

caused by the well’s proximity to the Longonot volcanic complex.

Stratigraphically, the Longonot pyroclastics (light grey to grey in colour) overlie the Olkaria

pyroclastics (yellowish brown in colour), attesting that the former are younger in age. Like

in most Olkaria wells, trachyte is still the dominant rock unit in the study well, in some cases

extending steadily downward to reach a thickness of about 1200 m (e.g. between 1700-2900

m). The unit forms the reservoir rock in the study well. Clearly, the rhyolites are markedly

less abundant in the study well relative to wells OW-905A, 910 and 917, perhaps caused by

the well's location being outside the main Olkaria volcanic zone. As a result, rhyolitic

volcanism seems to have been majorly confined to the proposed Olkaria caldera structure.

No intrusive rocks were intersected by the well. Intrusives result from the injection of magma

into the overlying country rock and commonly serve as a source of heat to shallow levels of

the geothermal system. Based on this observation, it is probable that pre- and post-caldera

magmatism associated with the OVC had minimal effect in the area surrounding the well.

Lack of intrusive rocks in the well further attests to the location of the well being outside the

main Olkaria volcanic zone.

5.2 Hydrothermal alteration

Hydrothermal alteration sequences provide a significant insight into the history of a

geothermal system. Hydrothermal alteration observed in the study well is essentially made

up of two components: (a) replacement of primary components in the rocks by alteration

minerals, and (b) precipitation of alteration minerals into open spaces (veins, fractures and

vesicles) in the rock. Generally, there is scarcity of high-temperature alteration minerals in

well OW-922. The top most zone of the well (0-190 m) contains relatively fresh rocks,

including a sequence that do not show any indication of geothermal interaction and at the

most show moderate oxidation due to groundwater circulation. The subsequent zone

between 190 and about 450 m is dominated by relatively low-temperature alteration mineral

assemblages (e.g. zeolites, smectites, chalcedony), signifying that inferred alteration

temperatures are < 200oC. Petrographic examination of the rock samples between this

interval indicate that the primary phases replaced are mostly volcanic glass and olivine, as

seen in Table 3. A prograde mineralogical change, marked by transition from secondary

quartz, wairakite, albite, illite, chlorite and epidote is evident between 476 m to about 1798

72

m depth. The occurence of epidote, which is regarded as one of the most important

greenschist alteration mineral assemblages (e.g. Bird and Spieler, 2004) is relatively

spasmodic. In most cases, the occurence of the mineral is observed as faint colouration,

possibly reflecting that equilibrium between the host rocks and reservoir fluids was seldom

achieved. It is worth noting that certain minerals, including calcite and pyrite form at both

low and high temperatures (i.e. between ~ 50-300oC), thus cannot be totally relied upon as

accurate temperature indicators (Brown, 1978). Below 1038 m depth to the well bottom,

mixed layer clays (MLCs), occuring in different combinations (smectite/illite,

smectite/chlorite and smectite/illite/chlorite), and calcite are the dominant alteration mineral

phases.

Clay mineralogy in the study well differs somewhat from clay alteration commonly observed

in other Olkaria wells. In particular, the persistence of MLCs to the bottom of the well

possibly reflects modification of the composition of hydrothermal fluids. This is in

agreement with reports of various authors (e.g. Steiner, 1977; Kristmannsdóttir, 1979) who

found out that the distribution of clay minerals in geothermal areas depends on the fluid

chemistry and thermal structure of the field. Studies of chemical composition of the fluids

are, however, required to prove the case. In previous studies in Olkaria (e.g. Mwania et al.,

2013; Musonye, 2015), it is observed that the occurence of MLCs is mostly restricted to

depths above 1500 m. MLCs in this well are characterized by well developed smectite peaks

at greater depths, for example at 2616 m, 2840 m and 2918 m depth, just to mention a few,

perhaps indicating that temperatures are not high enough. The association of MLCs with

haematite and abundant calcite, particularly towards the bottom of the well could reflect that

temperatures are relatively low in this region.

5.3 Permeability and comparison of

temperatures

Feed zones below the production casing in a well can either be cold or hot. Cold feed zones,

depending on the size, may ultimately cool the well, rendering it less productive. On the

other hand, hot feed zones in many cases are known to maximise well productivity. In the

study well, feed zones inferred demonstrate a relationship between geological observations,

variations in drilling parameters and kinks in temperature recovery logs. However, it is key

to point out that intrusion related feed zones, which are common in wells OW-905A, 910

and 917 are not present in well OW-922 as already discussed. Based on analysis of the feed

zones in relation to the indicators mentioned above, it is noted that most of the feed zones

are small, leading to the conclusion that permeability in the well is limited. This observation

is also supported, for example by the failed injection test where the well did not accept water,

absence of clear pivot point and total failure of the well to discharge after 49 days of heat-

up. The pivot point is fundamental in the sense that it is ordinarily used to infer the best feed

zone in a well. At this point, the pressure is nearly constant. Moreover, it is probable that

intense calcite deposition, particularly towards the bottom of the well, clogged the once

present primary and secondary fluid channels. It is also observed that a correlation between

secondary mineralisation and identified feed points is restricted, and the values of key

indicators of water/rock interaction (i.e. Na2O and K2O) do not show any large shift within

and in the vicinity of the feed zones as would be expected.

73

A comparison of measured formation and alteration temperatures (Figure 22) indicates that

the geothermal system around the study well has substantially been disturbed. It is apparent

that the inferred alteration temperature plots higher than the measured formation

temperature. This demonstrates that past temperatures were substantially higher than the

present measured ones. Temperature profile argues in favour of this observation, showing a

maximum temperature of 127oC betweeen ~2400- 2500 m depth (Figure 21), below which

there is a reversal in temperature gradient to the well bottom. This is inconsistent with the

inferred alteration temperature of about 240oC, denoted by the presence of epidote. It is

necessary though to mention that the measured formation temperature was taken only after

49 days recovery period. As a result, the formation temperature may not be representative

enough, since the geothermal system had likely not fully stabilized following the long

drilling time of the well. An up to date temperature log would provide a true reflection of

the current state of formation temperature in the well. It is, therefore, likely that significant

cooling has occurred in the geothermal system around well OW-922 over time.

Other possible indications of cooling in the well is demonstrated by the extensive deposition

of calcite (e.g. Figure 21), especially towards the well bottom (i.e. ~ 2240-2990 m). A

reasonable explanation for the intense calcite deposition, which is consistent with the

argument of Simmons and Christenson (1994), implies heating of the colder peripheral fluids

entering the geothermal system by the host rocks, consequently favouring carbonate

precipitation. This observation is also in accord with the interpretation put forward by

Musonye (2015) to explain intense calcite deposition in well OW-917. The latter was drilled

along the proposed caldera structure (Figure 3) and exhibits more or less similar

hydrothermal alteration features to well OW-922. Cases of calcite superimposed on epidote

(e.g. at 2782 m depth) has been observed by several authors (e.g. Mortensen et al., 2014;

Franzson et al., 2008) and interpreted to reflect carbonate deposition during cooling of the

system by influx of colder fluids into the reservoir. However, fluid inclusion studies would

be necessary to provide detailed insights about the temporal evolution of the geothermal

system around the study well.

Equally, haematite is observed in moderate amounts (i.e. at level 2 on the scale of 1-3)

between 2342 and 2982 m depth. In this case, the occurence of haematite can be interpreted

to suggest influx of oxygen-rich cooler groundwater into the system. The geothermal system

around the study well can therefore, be described as episodic rather than continuous. An

episodic geothermal system refers to a system where some changes (either heating or

cooling) has occured over time based on differences between inferred alteration and

measured (or calculated) formation temperatures. On the other hand, a continuous system is

where the inferred alteration and measured (or calculated) formation temperatures are more

or less in a state of equilibrium (Brown, 1978).

74

Figure 35: A possible conceptual model based on outcome of the study, along the

traverse A-B.

75

5.4 Rock composition

Rocks from the study well show a complete range of compositions from basalt to trachyte

or rhyolite. The dominance of evolved rock suites relative to mafic and intermediate rocks

implies high degree of differentiation of the parental magma. Peccerilo et al. (2007) observed

that the relative scarcity of the intermediate rocks are related to their high viscosity, which

inhibits them from erupting. Even though the more silicic magmas (pantellerites and

comendites) still have high viscosities, they tend to be associated with much lower density.

This feature, alongside the high volatile contents (e.g. CO2, H2O, Cl and F) of silicic

magmas, helps to drive them out after high degrees of fractional crystallisation from the

magma reservoirs (Ronga et al., 2009). Based on the TAS classification scheme, the

prevalence of trachytes is in accord with the general stratigraphy of Olkaria, which is

composed primarily of trachytes, except for the Olkaria west fields.

5.5 Hydrothermal alteration effects on rock

chemistry

Interaction of high-temperature fluids with host rocks (hydrothermal alteration) in the

geothermal system around well OW-922 has produced changes in the rocks. Essentially,

hydrothermal alteration is a very complex process involving a chemical, mineralogical and

textural changes of the host rock. The process results from the interaction of the wall rock

with hot aqueous fluids (Pirajno, 1992). These fluids deliver the chemical reactants and take

away the aqueous reaction products. The complex water-rock interaction process has

undoubtedly changed contents of the chemical elements in the altered rocks. In this case, a

direct method, based on comparison of elemental concentrations has been applied to

determine the effects of hydrothermal alteration on selected major and trace elements. Even

though this method serves the purpose to some extent, a distinct disadvantage is that it does

not put reliable constraints on the effect of alteration on element mobility in altered rocks.

Other methods, such as the statistical approach (e.g. Student t-tests) and mass balance

calculations, are known to yield more coherent results. Quantification of alteration effects

on an element is based on the concept of element mobility. The amount of hydrothermal

alteration is very often correlated with an element’s ionic potential (charge/radius ratio) (e.g.

Verma et al., 2005). Consequently, highly mobile elements are characterised by low ionic

potential (< 0.03 pm-1) and tend to be easily leached away by the geothermal fluids as

hydrated cations. These elements exhibit large variations (scatter) as a consequence of the

alteration process and are exemplified here by Na, K, Ba, Sr, and Rb. In contrast, most

immobile elements are characterised by high ionic potential (> 0.10 pm-1) and tend to be

retained in the host rocks since they are not easily leached away by hydrothermal fluids. In

this study, these elements include Ti, P, Zr, Y, Sc, La and the transition metals (Co, Cr, Ni,

V).

5.5.1 Effects on major element geochemistry

Selected major elements have been plotted against SiO2 to decipher alteration effects on

major element geochemistry (Figure 36). SiO2 has been used as the reference or independent

variable because it is a nearly conservative element (oxide) during hydrothermal alteration

processes. This view has been advocated by Verma et al. (2005), who have presented

76

convincing experimental and theoretical results. The authors suggest that SiO2 contents of

most geothermal waters are generally low (< 0.15%). As a result, the SiO2 contents of the

altered rocks should be more or less unaffected, and hence, the use of Si or SiO2 as a

conservative independent element or oxide for assessing alteration behavior of other

elements (or element ratios) is widely accepted.

The variation of selected major elements vs. SiO2 indicates that hydrothermal alteration has

had insignificant effect on Al2O3 and MnO in comparison with Na2O and K2O. Na2O and

K2O display some scattering, which is more pronounced in K2O and, are traditionally known

to be highly mobile during hydrothermal alteration (e.g. Verma et al., 2005). These elements

are the dominant constituents in alkali feldspar, which is a primary rock forming mineral in

trachytes and rhyolites. Consequently, this raises the possibility that, as the two elements are

highly mobile, they have been affected by post-magmatic alteration processes. The

preferential mobility of the alkalis compared to aluminium during hydrothermal alteration,

could also be responsible for the marked metaluminous character of a majority of the

samples in the study well. This is expressed by molar proportions Al2O3 > Na2O + K2O. The

cause of scattering of MnO is not intuitively obvious, but seems to be related to variable

degrees of fractionation as opposed to a true scatter. Another possible explanation could be

the result of background noise introduced by using analyses of highly porphyritic samples

(e.g. Wilson, 1989).

1

2

3

4

5

6

7

8

45 50 55 60 65 70 75

Na 2

O (

wt.

%)

6

8

10

12

14

16

18

Al 2

O3

(wt.

%)

0

2

4

6

8

45 50 55 60 65 70 75

K2O

(w

t.%

)

SiO2 (wt.%)

0,0

0,1

0,2

0,3

45 50 55 60 65 70 75

Mn

O (

wt.

%)

SiO2 (wt.%)

Figure 36: Variation diagrams showing the concentrations of Na2O, Al2O3, K2O and

MnO against SiO2.

77

5.5.2 Effects on trace element geochemistry

A plot of selected trace elements against Zr (Figure 37) have been used to assess the

alteration effects on trace element geochemistry. Zr has been used as an independent variable

owing to its large range of variation and being regarded as a highly immobile element in

volcanic rocks (Franszon et al., 2008).

From the selected trace element vs. Zr. plots, it is evident that hydrothermal alteration has

the least effect on Sr, Ni and Ba, but has severely mobilised Rb, as shown by large scattering.

La and Y are comparatively inert and not much affected by hydrothermal alteration

processes. In previous studies (e.g. Verma et al., 2005), the two elements have been

considered as typical fluid-immobile elements. Rb is an alkali metal and, is a major

constituent of alkali feldspars. The element is highly mobile and is easily leached from the

0

100

200

300

400

0 500 1000 1500 2000 2500 3000

Sr (

pp

m)

0

50

100

150

200

250

0 500 1000 1500 2000 2500 3000

Rb

(p

pm

)

0

100

200

300

400

500

0 500 1000 1500 2000 2500 3000

La (

pp

m)

0

20

40

60

80

100

120

0 500 1000 1500 2000 2500 3000

Ni (

pp

m)

0

500

1000

1500

2000

2500

0 500 1000 1500 2000 2500 3000

Ba

(pp

m)

Zr. (ppm)

0

100

200

300

400

0 500 1000 1500 2000 2500 3000

Y (p

pm

)

Zr. (ppm)

Figure 37: Variation diagrams showing the concentrations of Rb, Sr, La, Ni, Ba and Y

against Zr.

78

rocks by fluid components of the hydrothermal alteration process. Sr, even though normally

considered mobile, has been least affected by alteration.

It must be emphasised that the ICP-OES analysis did not include measurement of loss on

ignition (LOI). The latter has been used in many cases (e.g. Franszon et al., 2008) to monitor

alteration. Highly altered rocks tend to have high LOI values, whereas, the unaltered or

slightly altered rocks are characterised by low LOI values. In addition, Verma et al. (2008)

observed that high LOI are associated with low Na2O abundances, strongly indicative of Na

loss during alteration processes.

5.6 Magmatic differentiation processes

Magma differentiation or segregation processes are central to understanding the origin of

sequences of volcanic rocks exhibiting systematic variations in composition, despite being

obtained from a likely analogous source component. Insights into these processes are derived

from studies of lavas or intrusions exposed on the surface or obtained through drilling of the

sub-surface geological formations, as is the case in the present study. Importantly, the rocks

provide excellent opportunity of tracing changes in major and trace elements during

magmatic evolution. Models of magma differentiation, including magma mixing/mingling,

crustal contamination, partial melting and fractional crystallisation, have been tested by

many authors (e.g. Wilson, 1989; Omenda, 1997; Peccerillo et al., 2007).

To decipher the differentiation mechanisms for the sub-surface rocks in well OW-922, SiO2

was chosen as the abscissa or index of differentiation owing to its abundance and wide

concentration range (59-73 wt.%). Consequently, it can be used to observe the progress of

differentiation processes. In spite of the wide range of SiO2 contents, it is pertinent to note

that majority of the samples have silica values clustering between 63 and 73 wt.%. All the

major oxides and trace elements have been plotted against SiO2 on Harker diagrams (Figures

23 and 24) and the interpretations discussed below.

5.6.1 Major elements

Progressive depletion in major element oxides, such as FeO (total iron expressed as ferrous

oxide), MgO, CaO, TiO2 and P2O5 show curved to near-curved trends in the variation

diagrams. Whole-rock SiO2 contents are negatively correlated with these elements,

indicating that simultaneous fractional crystallisation process of mafic mineral assemblages

from a basaltic parental magma is a possible cause. The inverse trend observed between FeO

and MgO vs. SiO2 (Figure 23) can be interpreted as implying the separation of

ferromagnesian mineral phases (olivine and clinopyroxene) from the melt during early stages

of crystallisation (e.g. Harðarson, 1993). Similarly, the decrease in CaO content strongly

suggests that the mafic to intermediate magmas evolved by fractionation of an assemblage

dominated by plagioclase and clinopyroxene.

The interpretation of TiO2 vs. SiO2 could be related to fractionation of Fe-Ti oxides (e.g.

ilmenite and/or titano-magnetite) from the intermediate to silicic rock suites. According to

Wilson (1989), TiO2 is incompatible in the main minerals of basalt. As a result, its

concentration generally increases significantly in the residual liquid after crystallisation of

olivine and plagioclase during early stages of crystallisation. The decrease in P2O5 as SiO2

content increases is ascribed to the beginning of apatite crystallisation in the intermediate to

silicic rocks, resulting in a less phosphor content of the melt.

79

The down-turn in Na2O abundances in samples containing greater than 64 wt.% SiO2 (Figure

23) denotes the point at which plagioclase fractionation becomes a controlling factor in the

compositional evolution of the residual liquid. This observation is consistent with the gradual

decline in Al2O3 contents, which would only require fractionation of plagioclase, and to a

lesser extent alkali feldspar (Weaver et al., 1972). The CaO/ALK (ALK = total alkali) ratio

for the analysed samples decreases markedly as SiO2 content increases (Figure 22). This

trend is consistent with a fractional crystallisation model (Weaver et al., 1972). In general,

the different slope of trends for the major oxides vs. SiO2 is likely influenced by the

fractionating minerals and their proportions. Low P.I. (0.26-1.03) of the samples analysed

could probably be attributed to significant loss of the alkalis which are easily leached away

during hydrothermal alteration processes as discussed earlier. Based on this observation, it

is reasonable to suppose that the loss of Na2O and K2O in this well was sufficient enough to

make all but two samples (i.e. 1106 and 1142 m) appear metaluminous (i.e. Al2O3 > Na2O

+ K2O).

5.6.2 Trace elements

The relative enrichment and/or depletion of trace elements offers valuable insights about the

source characteristics or petrogenetic process of any volcanic rocks suite. It is important to

note that trace elements do not form any major rock forming minerals themselves but mostly

enter crystallographic sites in the most common mineral phases by substitution. Various

mineral phases may either accept (compatible) or largely exclude (incompatible in such

phases) these elements into their crystal structure. Consequently, the use of trace elements

to diagnose petrogenetic processes leads to better understanding of the nature and

composition of the mineral assemblages with which the magma may have previously

equilibrated. Although the trace elements graphics generally show more scattering (Figure

24) compared to the major element oxides (Figure 23), systematic variations with SiO2

contents are clearly evident.

The decrease in Co, Cr, Ni and V with increasing SiO2 content (Figure 24) can be

interpreted to signify fractional crystallisation of mafic mineral assemblages (olivine and

clinopyroxene) from a basaltic parental magma. From the description of the lithologies in

well OW-922, it is noticeable that olivine and clinopyroxene are frequently observed as the

dominant phenocryst phases in majority of the mafic rocks. Nickel partitions easily into

olivine while V is mainly incorporated into clinopyroxene in mafic rocks (Wilson, 1989).

The depletion of Sc towards more evolved rocks is consistent with the fractionation of

olivine and plagioclase mineral phases. Cobalt follows similar pattern as Sc and is known

to partition strongly into olivine (Weaver et al., 1972). Extraction of Cu is usually not

related to any precipitating major mineral phases but can be linked to the separation of a

sulphide minerals (Weaver et al., 1972).

The main phase affecting Sr partitioning in the mafic rocks is plagioclase and to a lesser

extent alkali feldspar. Therefore, the systematic enrichment in Sr contents in the mafic and

intermediate rocks can be accounted for by the accumulation of plagioclase (and possibly

alkali feldspar). Additionally, incorporation of Sr into apatite or titanite may as well

contribute to its reduction in the more evolved lavas (Weaver et al., 1972). However, most

of the trend observed in Sr. vs. SiO2 plot can be attributed to variable degrees of plagioclase

fractionation. ITEs, e.g. Y, Zr, Rb and La, show a relatively good positive correlation with

increasing SiO2 contents despite some scattering (Figure 24). The correlation in Rb vs. SiO2

can be termed strong correlation, whereas for Y, Zr and La vs. SiO2 is a moderately strong

correlation. Increase in ITEs concentrations with increasing SiO2 content can satisfactorily

80

be interpreted to mark their incompatible behavior, where the melt become steadily

enriched in these elements as fractionation proceeds.

The strong depletion of Ba (< 20 ppm) in rhyolites can be attributed to fractionation of

sanidine. The element becomes compatible when alkali feldspar appears in the phase

assemblage of intermediate and silicic melts. Here, it easily substitutes for potassium in K-

feldspar, specifically the sanidine. The latter is one of the prominent phenocryst phases in

the rhyolites. It is important to mention that fractionation of sanidines also ultimately

affects the concentration of potassium in the residual liquid. On the other hand,

accumulation of sanidines could possibly account for Ba enrichment in the trachytes. The

differentiation trends seen in the major and trace element systematics (Figures 23 and 24)

are consistent with a fractional crystallisation process for the evolution of the rocks in well

OW-922.

5.6.3 Assessing other differentiation mechanisms

To assess other petrogenetic processes responsible for the generation of these rocks besides

fractional crystallisation, plots of incompatible vs. incompatible elements are shown below.

Zr is used as an independent variable owing to its large range of variation, high

incompatibility and relatively immobile nature during hydrothermal alteration (Franzson

et al., 2008).

As discussed earlier, fractional crystallisation is the most viable model for the generation

of the evolved rocks in this well. This has been emphasized further by the La and Y vs. Zr

plots. A basic requirement for fractional crystallisation processes for any two ITEs is that

the two elements must plot on linear trends which can be extrapolated to the origin

(Omenda, 1997). In his pronouncement, fractional crystallisation does not significantly

change ratios of trace elements with analogues degrees of incompatibility. This

requirement has been successfully fulfilled by La and Y vs. Zr plots. However, fractional

crystallisation cannot be single-handedly put forth to explain all the geochemical features

depicted by La and Y vs. Zr plots.

0

100

200

300

400

500

0 500 1000 1500 2000 2500 3000

La (

pp

m)

Zr. (ppm)

0

100

200

300

400

0 500 1000 1500 2000 2500 3000

Y (p

pm

)

Zr. (ppm)

Figure 38: A plot of La vs. Zr. and Y vs. Zr. showing different magmatic differentiation

processes. The dotted and continuous lines indicates fractional crystallisation and other

petrogenetic processes, respectively.

81

Deviations from the continuous trends in the incompatible element plots (La/Zr, Y/Zr)

observed in the rhyolites suggest that other processes (e.g. through partial melting of crustal

rocks or magma mixing) at least contribute to their petrogenesis. These deviations are

exhibited predominantly by the rhyolites between 1106-1142 m and 2420-2570 m depth.

Previous studies of the magmatic differentiation mechanisms for the surface rhyolites in

Olkaria (e.g. Macdonald et al., 2008) agree with the observation made in the present study.

In their study, the authors established that rhyolites were formed primarily by fractional

crystallisation of metaluminous trachytic magmas and to a lesser extent by partial melting

of syenites. The trachytes themselves, were formed by the fractional crystallisation of

basaltic magmas through mugearites and benmoreites (Marshall et al., 2009).

It is therefore possible that rhyolites plotting off the main trend were generated by partial

melting of the syenites. However, no clear theory based on geochemical grounds has been

put forward by the authors to exhaustively explain the origin of rhyolites by partial melting,

even though there is a high probability that the process was promoted by fluxing of volatiles

emanating from underlying magmas. Evaluating the degree of partial melting is, however,

not feasible, because doing so requires knowledge of partition coefficients for the relevant

mineral phases, source compositions and the depth at which melting occurs. These are

rather poorly constrained.

5.7 Comparison of surface and sub-surface

samples

A comparison of surface samples from Olkaria and neighboring minor eruption centres and

sub-surface samples from the study well, and wells OW-905A, 910 and 917 (Figure 27)

clearly points towards a common petrogenetic mechanism for these rocks. This is exhibited

by both samples plotting within similar fields on the TAS diagram (Figure 27). On this

account, it is reasonable to suggest that magma from these volcanic centres was generated

by a common differentiation process, possibly from a similar parental magma. Only detailed

radiogenic isotope studies can convincingly prove the case. Correspondingly, comparison of

the geochemical data for selected major and trace elements for the four wells (Figures 28

and 29) demonstrate a perfect overlap across the whole range of compositions. A more

realistic interpretation would be that the trends displayed by the two groups of elements are

related by a common evolution mechanism.

These results are in agreement with the interpretations of Musonye (2015), who presented

convincing results in his study of the petrochemistry of wells OW-905A, 910 and 917. The

author construed high degree of correlation between certain major and trace elements vs. an

evolution index as being consistent with predominantly extended fractional crystallisation,

or even requiring fractional crystallisation as a necessary condition. Since the rocks from

well OW-922 are geochemically comparable with rocks from wells OW-910, 917 and 905A,

a petrogenetic connection seems likely between the rocks in the four wells.

Whilst these relationships are consistent with a fractional crystallisation process, it is not

necessarily the only mechanism that accounts for all the geochemical characteristics

observed. Other differentiation processes cannot be completely ruled out, however, limited

role they play in the evolution of these rocks. To illustrate this point, plots of two ITEs

plotted against each other have been made, as in Figure 38. Results of these plots (Figure

39) coincide with interpretations made for Figure 38. Once again, the unanswered question

82

remains whether these rocks are derived from similar or different parental magma. Once

again, only detailed radiogenic isotope studies can resolve the impasse.

5.8 Compositional variation of alteration

minerals

5.8.1 Epidote

Measured chemical compositions of different epidote crystals, derived from two different

depth intervals in well OW-910, show wide and concomitant variations in Al and Fe

contents. Variations in the Fe3+-Al3+ contents of epidote observed presumably reflect

changes in fluid composition during the formation of the mineral. Epidote composition

depends directly on the ratio of the activities of Fe3+and Al3+ in the fluid. However, aqueous

complexation of Fe3+ and Al3+ with other anions (e.g. SO42−, CO3

2−, Cl−, OH−, etc) may cause

the composition of epidote to depend on ƒCO2, pH and chloride concentration in the fluid

phase (Arnason et al., 1993). Thus, changes in fluid composition during the evolution of

epidote may result in variations in Fe3+and Al3+ exchange without necessarily reflecting

changes in temperature or pressure.

Other plausible thermodynamic mechanisms to account for the variation in epidote

composition in well OW-910 are changes in ƒO2 and the oxidation state. Diverse

experimental results (e.g. Arnason et al., 1993) have demonstrated that ƒO2 strongly controls

the Fe-content of epidote. Values for ƒO2 for the Olkaria geothermal system have not been

determined, but such studies would be of great interest. According to the authors, epidote is

the most Fe-rich in the presence of haematite-magnetite oxygen buffer and becomes more

aluminous with decreasing ƒO2, as the proportion of ferric iron in the system decreases. In

addition, the authors noted that epidote stability increases with increasing ƒO2. The latter

expands the P-T boundaries of the epidote stability field. Like in the Salton Sea geothermal

system, epidote in well OW-910 is known from binocular observation and optical

petrography of cuttings to occur alongside haematite. This observation is closely compatible

0

100

200

300

400

500

0 1000 2000 3000

La (

pp

m)

Zr (ppm)

OW-922 OW-905A OW-910 OW-917

0

100

200

300

400

500

0 500 1000 1500 2000 2500 3000

Y (p

pm

)

Zr (ppm)

OW-922 OW-905A OW-910 OW-917

Figure 39: Harker's variation diagrams between various pairs of incompatible trace

elements displaying fractional crystallisation (dotted line). The flexure from the main

trend (continuous line) represent other magma differentiation processes.

83

with the relatively high ƒO2 calculated for the Salton sea geothermal system (Bird and

Spieler, 2004). Epidote in haematite-bearing assemblages is traditionally known to be more

Fe-rich (Bird et al., 1988).

Additionally, results of epidote analysis at 1236 m depth display a positive relationship

between the relatively high Fe-content of epidote and inferred level of oxidation. This is

marked by concomitant increase in the oxidation state of host rocks from 864 m to 1236 m

depth. The enrichment of Sr and Ti in the rim and core of epidote, respectively suggest that

epidote could be an important repository for these elements, mobilized through hydrothermal

alteration. Results of previous analyses of Icelandic epidotes, for example in the Hengill

geothermal system (e.g. Bragason, 2012), points to a more or less similar picture where, Sr

contents are noted to increase progressively towards the rims. Based on the authors'

interpretation, increase in Sr contents very likely reflects extensive breakdown of

plagioclase.

Variations of the average mole fraction of pistachite (Xps) contents of epidote in this well is

satisfactorily correlated with depth, and is in agreement with the range in some well known

geothermal systems of the world (e.g. Bird and Spieler, 2004, Bird et al., 1988). Marks et al.

(2010), while studying epidote in the Reykjanes and Nesjavellir high-temperature

geothermal systems in Iceland observed an increase in the average molar ratio of pistachite

with depth. The authors interpreted this observation to indicate an increase in temperature.

In the Olkaria geothermal system, well OW-910 is a high-temperature well with regard to

measured down-hole formation temperatures, reaching a maximum of about 300oC

(Musonye, 2015). The distribution of formation temperature in this well, therefore, is

consistent with the relationship between the molar ratio of pistachite and temperature

advocated by Marks et al. (2010). Temperature exerts a major influence on the transition of

silicate phases, and consequently controls the phase equilibrium (Arnason et al., 1993). By

contrast, this correlation is contradictory with the results of epidote analyses from the Salton

Sea geothermal system by Bird et al. (1988). Based on their observations, the authors

established that epidote becomes more aluminous with increasing depth and temperature.

Both changes in fluid composition and temperature were attributed as the most influential

variables causing changes in epidote composition in that case.

Zoning of epidote mineral grains observed, for example by Marks et al. (2010) in the

Reykjanes and Nesjavellir geothermal systems in Iceland is also confirmed in the present

study (e.g. Figure 32). The explanation for compositional zoning patterns in epidote is an

intriguing problem. This is not suprising considering the sensitivity of Fe and Al contents of

epidote to slight changes in a variety of intensive and extensive thermodynamic variables

during hydrothermal activity. Zoning also depends directly on the initial ratio of the activities

of Fe3+ and Al3+ in the aqueous solution. In well OW-910, intragrain compositional zoning

seems to be dependent directly on local changes in fluid transport properties and intensive

thermodynamic variables (e.g. CO2 concentrations in fluid phase, pH, bulk rock composition

and ƒO2) during the thermal evolution of the geothermal system. Such changes result in the

formation of several spatially related phases which are not in chemical equilibrium. In this

respect, it is reasonable to assume that intrusion activity (e.g. basaltic and micro-granite)

observed in this well caused a surge in CO2 concentrations in the thermal fluids. This in turn

caused a change in the fluid composition by lowering the pH of the fluids, consequently

affecting aqueous speciation of Al3+ and Fe3+ complexes. While this is important in many

cases, compositional zoning patterns in some epidote mineral grains are caused by non-

equilibrium, irreversible processes (Arnason and Bird, 1992). In this case, the relative rates

84

of Al3+ and Fe3+ mass transfer accompanying the irreversible dissolution of common rock

forming minerals control epidote compositional zoning.

5.8.2 Chlorite

The Fe-rich nature of chlorites seems to be correlated to the relatively high Fe content of the

bulk rocks, as seen from the results of whole-rock analysis from wells OW-922 (Table 5)

and 910 (Musonye, 2015). This observation is however, not valid for the relatively Mg-rich

chlorites at 864 m depth from well OW-910. The most likely explanation for these

differences is the structural limitation in the assignment of Fe2+ into trioctahedral chlorite,

which favours the incorporation of Mg2+ (rather than Fe2+) under oxidizing conditions, as

demonstrated by Bryndzia and Scott (1987). If a Mg-rich solution, such as seawater-

dominated hydrothermal solution, had its way into a geothermal system, chlorite would be

expected to be Mg-rich. The relatively low quantities of Na, K and Ca probably rule out the

presence of corrensite layers in the analysed chlorite from Olkaria geothermal system.

As a result of the difficulty in obtaining precise chlorite analyses, occasioned by the

relatively porous and uneven surface, it was not feasible to draw any conclusions about the

relationship between chlorite composition and formation permeability. Results of the molar

ratio of Fe2+ / (Fe2+ + Mg) in chlorite presented in Table 9 are inconsistent with the

observation made by Lonker and others (1993) for wells RN-8 and 9 in the Reykjanes

geothermal system, Iceland. In their study, the authors reported well defined correlation

between chlorite composition and formation permeability. They observed a progressive

increase in Fe2+ / (Fe2+ + Mg) ratio and a decrease in Si contents within or in the vicinity of

aquifers in wells RN-8 and 9. Reverse correlations were however noted from the less

permeable formations. In well OW-922, 2236 m depth coincides with a small feed zone

(Table 4), while 1236 m depth in well OW-910 is slightly below a major feed zone

(Musonye, 2015).

85

6 Conclusion and Recommendations

6.1 Conclusion

The lithological units encountered in well OW-922 are consistent with the general sub-

surface stratigraphy of the Olkaria Volcanic Complex. Total absence of intrusive rocks and

a marked reduction in rhyolitic volcanism is, however, observed in this well. The two aspects

point towards the location of the well being outside the main Olkaria volcanic zone.

Generally, the well is dominated by low-temperature alteration minerals, which are observed

extending to greater depths. Abundant calcite deposition is apparent, especially towards the

bottom of the well, and the mineral is observed overprinting epidote.

Five small feed zones have been recognized in the well. They are connected to lithological

contacts, variations in alteration patterns, circulation losses, prevalence of calcite deposition,

intense pyritization, changes in ROP and kinks in temperature recovery logs.

Intense calcite deposition, failure of the well to accept water during injection test and the

absence of clear pivot point are indications that permeability in the well is very limited.

Hydrothermal alteration mineralogy shows that the area has had temperatures ≥ 240oC in the

past, as indicated by the presence of epidote. However, measured formation temperature

reached a maximum of 127oC after 49 days recovery period. Consequently, the difference in

measured formation and inferred alteration temperatures distinctly demonstrates that the

geothermal system around the well has significantly cooled over time. Calcite overprinting

epidote further argues in favour of this conclusion.

From petrochemical analysis, the rocks show a complete range of compositions from basalt

to trachyte or rhyolite. Highly evolved rocks are predominant.

Based on major oxides and trace elements systematics, fractional crystallisation is the

dominant magma differentiation mechanism for evolution of the rocks of the study well.

Other magma modification mechanisms likely played some role, but only to a limited extent.

Significant changes in the abundance of major oxides and trace elements have not taken

place during hydrothermal alteration processes. Equally, a comparison of surface and sub-

surface samples indicates the possibility of a petrogenetic connection caused by a common

differentiation process. The samples are, therefore, geochemically comparable.

The compositional variations of epidote from two different depth intervals in well OW-910

appear to be strongly influenced by the combined effects of temperature, ƒO2 and fluid

composition. Results of similar studies from other geothermal systems corroborates this fact.

The presence of zonation in epidote mineral grains indicates that fluid composition and

pressure or temperature conditions in the geothermal system have changed with time.

Due to limited number of epidote and chlorite samples, it was not possible to carry out a

stratigraphic correlation in the four wells based on major and minor element contents.

86

6.2 Recommendations

Additional temperature and pressure profiles need to be measured. As the well has been shut-

in for a period of over one and a half years, current temperature and pressure measurements

will provide a true reflection of the state of the system, as opposed to the 49 days recovery

profile used in this study.

Fluid inclusion studies need to be carried out and integrated with the inferred alteration and

measured (or calculated) formation temperatures. Several attempts were made to conduct

fluid inclusion analysis for the well but were unsuccessful due to problems with sample

preparation. The findings would serve as a better means of reconstructing the past history of

the geothermal system around well OW-922.

In-depth geological and geophysical studies of the area east of Domes field is still strongly

recommended if expansion of production drilling has to continue in the area. Some

geophysical studies have already been carried in the area but the findings are limited in my

view. The combination and comparison of the findings with other disciplines (i.e.

geochemistry) will be crucial in better understanding the system east of the Domes.

Well OW-922 should be listed amongst the reservoir monitoring wells in Olkaria. In

particular, monitoring of the drawdown changes in this well will be vital in establishing

whether or not there exists a connection between system around the well and the Olkaria

reservoir.

There is need to carry out radiogenic isotope studies (Pb-Sr-Nd-Hf) to gain valuable

information about the source of the rocks from the four wells, that is whether the rocks are

derived from similar or different parental magmas. In addition, radiogenic isotopes will

provide useful insights about crustal contamination during ascent of the magma.

87

References

Achauer, U. and Masson, F., 2002: Seismic tomography of continental rifts revisited: from

relative to absolute heterogeneities. Tectonophysics, 358, p. 17-37.

Allen, D.J., Darling, W.G., and Burgess, W.G., 1989: Geothermics and hydrogeology of the

southern part of the Kenya Rift Valley with emphasis on the Magadi-Nakuru area. British

Geological Survey, report SD/89/1, 68 pp.

Ambusso, W.J., and Ouma, P.A., 1991: Thermodynamic and permeability structure of

Olkaria Northeast field: Olkaria fault. Geothermal Resource Council Transactions, 15, p.

237-242.

Arnason, J.G., and Bird, D.K., 1992: Formation of zoned epidote in hydrothermal systems.

Water-Rock interaction, Stanford University, USA, p. 1473-1476.

Arnason, J.G., Liou, J. G, and Bird, D.K., 1993: Variables controlling epidote composition

in hydrothermal and low-pressure regional metamorphic rocks. Proceedings Abhand Geol.

Bund., Salzburg, Austria vol. 49, p. 17–25.

Baker, B.H., Mohr, P.A., and Williams, L. A. J., 1972: Geology of the Eastern Rift System

of Africa. Contrib. GSA Special Papers. Vol. 136, p. 1-68.

Baker, B.H. and Wohlenberg, J., 1971: Structural evolution of the Kenya Rift Valley. Nature

229, p. 538-542.

Bird, D. K. and Spieler, A. R., 2004: Epidote in Geothermal systems. Reviews in Mineralogy

and Geochemistry, Vol. 56, p. 235-300.

Bird, D.K., Cho, M., Janik, C.J., Liou, J.G., and Caruso, L.J., 1988: Compositional,

order/disorder, and stable isotope characteristics of Al-Fe epidote, State 2-14 Drill hole,

Salton Sea geothermal system. J. of geophysics. Res., 93, p. 13135-13144. Bosworth, W., 1987: Axis volcanism in the Gregory Rift, East Africa: implications for

models of continental rifting. Geology, 15: p. 397-400.

Bliss, C.M. 1979: Geology and petrochemistry of the Naivasha area, Kenya RiftValley.

Ph.D. thesis (unpublished), Lancaster University, p. 1-270.

Bragason, G, O., 2012: Strontium isotope shift in clay minerals, epidote and geothermal fluid

in the Hellisheiði Geothermal Field, SW-Iceland, University of Iceland, MSc. Thesis, p. 1-

45.

Browne, P.R.L., 1984: Lectures on geothermal geology and petrology. UNU-GTP, Iceland,

report 2, 92 pp.

Browne, P.R.L., 1978: Hydrothermal alteration in active geothermal fields. Annual Review

Earth and Planetary Sciences, 6, p. 229–250.

88

Bryndzia, L.T. and Scott, S.D., 1987: The composition of chlorite as a function of sulfur and

oxygen fugacity. An experimental study. Am. J. Sci., 287: p. 50-76.

Chorowicz, J., 2005: The East African Rift System. Journal of African Earth Sciences, 43,

p. 379-410.

Clarke, M.C.G., Woodhall, D.G., Allen, D., Darling, G., 1990: Geological, Volcanological

and hydrogeological controls of the occurrence of geothermal activity in the area

surrounding Lake Naivasha, Kenya. Min. of Energy, Nairobi, report 150, 138 pp. Davies, G.R., and Macdonald, R., 1987: Crustal influences in the petrogenesis of the

Naivasha basalt-comendite complex, combined trace elements and Sr-Pb isotope constraints.

Journal of petrology 28, p. 1009-1031.

Dubois, M., and Martinez-Serrano, R, G., 1998: Chemical variations in chlorite at the Los

Humeros geothermal system, Mexico. Clay and clay minerals, Volume 46, No. 6, p. 615-

628.

Ebinger, C., 2005: Continental breakup: the East African perspective. Astronomy and

Geophysics, 46, p. 216–221.

Ebinger, C. and Sleep, N. H., 1998: Cenozoic magmatism in central and east Africa resulting

from impact of one large plume. Nature 395, p. 788–791.

Franzson, H., Zierenberg, R. and Schiffman, P., 2008: Chemical transport in geothermal

systems in Iceland, Evidence from hydrothermal alteration. Journal of Volcanology and

Geothermal Research, Volume 173, p. 217-229.

Harðarson, B.S., 1993: Alkalic rocks in Iceland with special reference to the Snaefellsjokull

Volcanic System, Ph.D. thesis (unpublished), University of Edinburgh, p. 1-330.

Heumann, A., Davies, G.R., 2002: U-Th disequilibrium and Rb-Sr age constraints on the

magmatic evolution of peralkaline rhyolites from Kenya. Journal of Petrology 43, p. 557-

577.

KenGen 2014: Well completion report for well OW-922. KenGen in-house report

(unpublished), p. 1-32.

KenGen 2012: Stratigraphy and hydrothermal alteration mineralogy of well OW-915B.

KenGen internal report (unpublished), p. 1-40.

Kristmannsdóttir, H., 1979: Alteration of basaltic rocks by hydrothermal activity at 100-

300°C. In: Mortland, M.M., and Farmer, V.C. (editors), International Clay Conference1978.

Elsevier Scientific Publishing Co., Amsterdam, p. 359-367.

Laird, D.A., 2006: Influence of layer charge on swelling of smectites. Applied Clay Science,

34, p. 74-87.

89

Lagat, J.K., 2004: Geology, hydrothermal alteration and fluid inclusion studies of the Olkaria

Domes geothermal field, Kenya. University of Iceland, MSc. Thesis, UNU-GTP, Iceland,

report 2, 71 pp.

Le Bas, M.J., Le Maitre, R.W., Streckeisen, A., and Zanettin, B., 1986: A chemical

classification of volcanic rocks based on the total alkali-silica diagram: Journal of Petrology,

v. 27, no. 3, p. 745-750.

Lonker, S.W., Franzson, H., & Kristmannsdottir, H., 1993: Mineral-Fluid Interactions in

the Reykjanes and Svartsengi Geothermal Systems, Iceland. American Journal of Science,

293 (7), p. 605-670.

Macdonald, R., Black, H.E., Fitton, J.G., Marshall, A.S., Nejbert, K., Rodgers, N.W., and

Tindle, A.G., 2008: The roles of fractional crystallisation, magma mixing, crystal mush

remobilization and volatile-melt interactions in the genesis of a young basalt-peralkaline

rhyolite suite, the Greater Olkaria volcanic complex, Kenya Rift Valley. J. Petrology, 49-8,

p. 1515-1547.

Macdonald, R., Scaillet, B., 2006: The central Kenya peralkaline province: Insights into the

evolution of peralkaline salic magmas. J. Petrology, 2006, p. 59-73.

Macdonald, R., Rogers, N. W., Fitton, J. G., Black, S., and Smith, M., 2001: Plume-

Lithosphere interactions in the generation of basalts of the Kenya Rift, East Africa. Journal

of Petrology, vol. 42, p. 877-900.

Macdonald, R., 1974: Nomenclature and petrochemistry of the peralkaline oversaturated

extrusive rocks. Bull Volcanol 38: p. 498–516.

Mannvit/ÍSOR/Vatnaskil/Verkís Consortium, 2011: Consultancy services for geothermal

optimization study of the Greater Olkaria Geothermal Fields, Report 3: Revision of the

conceptual model of the Greater Olkaria Geothermal System – Phase I. Mannvit/ÍSOR/

Vatnaskil/Verkís, Reykjavík, November 2011, 100 pp.

Marks, N., Schiffman, P., Zierenberg, R.A., Franzson, H., and Fridleifsson, G.Ó., 2010:

Hydrothermal alteration in the Reykjanes geothermal system: Insights from Iceland deep

drilling program well RN-17. J. of Volcanology & Geothermal Research 189 (2010), p. 172–

190.

Marshall, A.S., Macdonald, R., Rogers, N.W., Fitton, J.G., Tindle, A.G., Nejbert, K., and

Hinton, R.W., 2009: Fractionation of peralkaline silicic magmas: the Greater Olkaria

volcanic complex, Kenya Rift Valley. J. Petrology, 50, p. 323-359.

Mielke, P., Nehler, M., Bignall, G., and Sass I., 2015: Thermo-physical rock properties and

the impact of advancing hydrothermal alteration- A case study from the Tauhura geothermal

field, New Zealand. Journal of Volcanology and Geothermal Research, vol. 301, p. 14-28.

Moore, M.D. and Reynolds, R. C. Jr., 1997: X-ray diffraction and the identification and

analysis of clay minerals. New York: Oxford University Press,18th ed., 378 pp.

90

Mortensen, A.K., Egilson, T., Gautason, B., Arnadottir S., and Gudmundsson A., 2014:

Stratigraphy, alteration mineralogy, permeability and temperature conditions of well IDDP-

1, Krafla, NE-Iceland. Geothermics, 49, p. 31-41.

Mungania, J., 1999: Geological report of well OW-714. Kenya Power Company internal

report.

Musonye, X.S., 2015: Sub-surface petrochemistry, stratigraphy and hydrothermal alteration

of the Domes area, Olkaria geothermal field, Kenya, University of Iceland, MSc. Thesis, p.

1-109.

Mwania, M.M., Munyiri, S., and Okech, E., 2013: Borehole geology and hydrothermal

mineralisation of well OW-35, Olkaria East geothermal field, Kenya, Report Number 1, p.

1-56.

Mwangi, D.W., 2012: Borehole geology and hydrothermal mineralisation of well OW-916,

Olkaria Domes geothermal field, Naivasha, Kenya. Reports 2012, Number 24, p. 541-571.

Nkapiani, M.N., 2014: Application of stable isotope geochemistry to tracing recharge and

flow systems of fluids in the Olkaria geothermal field, Kenya. Reports 2014, Number 24, p.

483-504.

Nyblade, A.A. and Robinson, S.W., 1994: The African superswell. Geophysical Research

Letters, 21, p. 765–768.

Ofwona, C., Omenda, P., Mariita, N., Wambugu, J., Mwawongo, G., and Kubo, B., 2006:

Surface geothermal exploration of Korosi and Chepchuk prospects. KenGen internal report,

44 pp.

Ojiambo, B.S., and Lyons, W. B., 1993: Stable isotope composition of Olkaria geothermal

field fluids, Kenya. Geothermal Resource Council, Transactions 17, p. 149-145. Okoo, J.A., 2013: Borehole geology and hydrothermal mineralisation of well OW–39A,

Olkaria East geothermal field, Naivasha, Kenya. Reports 2013, Number 24, p. 547-574.

Omenda, P.A., 1998: The geology and structural controls of the Olkaria geothermal system,

Kenya. Geothermics, 27-1, p. 55-74.

Omenda, P.A., 1997: The Geochemical Evolution of Quaternary Volcanism in the South-

Central Portion of the Kenya Rift: Ph.D. thesis, University of Texas at El Paso, p. 1-219.

Otieno, V., Okoo, J., Munyiri., S., Wanjohi, D., and Wanjohi A., 2014: Structural Mapping

of Olkaria South West Field, Olkaria Geothermal Project: KenGen Internal Report, 41 pp.

Otieno, V., and Kubai, R., 2013: Borehole geology and hydrothermal mineralisation of well

OW-37A, Olkaria east geothermal field, Kenya, Report Number 2, p. 57-105.

Pirajno, F., 1992: Hydrothermal Mineral Deposits. John Wiley and Sons, Sydney, 709 pp.

91

Peccerillo, A., Donati, C., Santo, A. P., Orlando, A., Yirgu, G., and Ayalew, D., 2007:

Petrogenesis of silicic peralkaline rocks in the Ethiopian rift: Geochemical evidence and

volcanological implications. Journal of African Earth Sciences 48, p. 161-173.

Reyes, A.G., 2000: Lectures on petrology and mineral alteration in hydrothermal systems:

From diagenesis to volcanic catastrophes. UNU-GTP, Iceland, report 18-1998, 77 pp.

Ring, U., 2014: The East African Rift System: Austrian Journal of Earth sciences, Vienna

2014. Vol. 107/1, p. 132-146.

Rollinson, H.R., 1995: Using geochemical data: Evaluation, presentation, interpretation.

Longman Group Ltd., England.

Ronga, F., Lustrini, M., Marzoli, A., and Melluso, L., 2009: Petrogenesis of a basalt-

comendite-pantellerite rock suite: the Boseti Volcanic Complex (Main Ethiopian Rift).

Miner Petrol, 98, p. 227-243.

Ronoh, I.J., 2012: Borehole geology and hydrothermal mineralisation of well OW-912B,

Olkaria geothermal field, Kenya. Report 29, p. 695-732.

Schiffman, P. and Roeske, S., 2014: Electron Microprobe Analysis of Minerals, University

of California, p. 293-306.

Shackleton, R.M., 1996: The final collision between East and West Gondwana: where is it?

Journal of African Earth Sciences, 23, p. 271-287.

Simiyu, M.S., 2010: Status of Geothermal Exploration in Kenya and Future Plans for its

Development. Proceedings World Geothermal Congress, 25-29 April 2010.

Simiyu, S.M., and Keller, G.R., 2000: Seismic monitoring of the Olkaria Geothermal area,

Kenya Rift valley, Journal of Volcanology and Geothermal Research 95, p. 197-208.

Simmons, S.F. and Christenson, B. W., 1994: Origins of calcite in boiling geothermal

systems. American Journal of Science, Vol. 294, p. 361-400.

Smith, M., Mosley, P., 1993: Crustal heterogeneity and basement influence on the

development of the Kenya Rift, East Africa. Tectonics 12 (2), p. 591–606.

Steiner, A., 1977: The Wairakei Geothermal Area, North Island, New Zealand: its sub-

surface geology and hydrothermal rock alteration. N. Z. Geol. Surv. Bull. 90, 127 pp.

Steiner, A., 1968: Clay minerals in hydrothermally altered rocks at Wairakei, New Zealand.

Clay and clay minerals, Vol. 16, p. 193-213.

Verma, S.P., Torres-Alvarado, I. S., Pandarinath, K., and Dulski, P., 2008: Element mobility

during the hydrothermal alteration of rhyolitic rocks of the Los Azufres geothermal field

(Mexico): Geothermic 37, p. 53-72.

92

Verma, S.P., Torres-Alvarado, I. S., Satir, M., and Dobson, P. F., 2005: Hydrothermal

alteration effects in geochemistry and Sr, Nd, Pb and O isotopes of magmas from the Los

Azufres geothermal field (Mexico): A statistical approach, Geochemical journal, Vol. 39, p.

141-163.

Weaver, B. D., Sceal, J.S.C and Gibson, I. L., 1972: Trace element data relevant to the origin

of trachytic and pantelleritic lavas in the East African Rift System. Contrib. Mineral. Petrol.

36, p. 181-194.

Williams, L.A.J., 1970: The volcanics of the Gregory Rift Valley, East Africa. Bulletin of

Volcanology, Vol. 34, p. 439-465. Wise, W.S., 2005: Zeolites. University of California-Santa Barbara, Santa Barbara, CA,

USA, Elsevier Ltd., p. 591-600.

Wilson, M., 1989: Igneous petrogenesis, A global tectonic approach. Department of Earth

Sciences, University of Leeds, p. 1-466.

Yoshio, W., Eiichiro, M. and Kozo, S., 2011: X-ray Diffraction Crystallography. New York:

Springer Heidelberg Dordrecht. p. 67-80.

93

Appendix A: Lithological description of well OW-922 based on binocular and

optical petrographic analyses.

0-46 m Pyroclastics: Light grey, unconsolidated heterogenous drill chips composed

predominantly of tuffaceous, pumiceous, obsidian, volcanic glass as well as obsidian. It is

moderately oxidized and no hydrothermal alteration is observed in this section. Alteration

minerals: oxides.

46-58 m: No samples (Loss of circulation returns).

58-96 m Pyroclastics: Light grey, considerably unconsolidated heterogenous drill cuttings

composed of tuffaceous, pumiceous, volcanic glass as well as obsidian. It is moderately

oxidized and no hydrothermal alteration is observed in this section. Alteration minerals:

oxides.

96-190 m: No samples (Loss of circulation returns).

190-302 m Pyroclastics: Light grey to yellowish-brown, unconsolidated heterogenous drill

chips composed predominantly of tuffaceous, pumiceous, volcanic glass as well as obsidian.

It is mildly oxidized. Alteration minerals: oxides, calcite.

302-350 m Tuffs (lithic): Brown to light grey, cryptocrystalline tuffaceous rock. It is highly

vesicular, vesicles are primarily infilled with zeolites (scolecite and mesolite) and rarely

calcite. Patches of fine grained green clays are also noted infilling vesicles. The rock is

weakly altered with minor pyrite disseminations in the matrix. Alteration minerals: clays,

calcite, zeolites.

350-362 m: No samples (Loss of circulation returns).

362-364 m Tuffs: Brown to light grey, cryptocrystalline tuffaceous rock. It is highly

vesicular, vesicles are primarily infilled with zeolites (scolecite and mesolite) and rarely

calcite. The rock is weakly altered with minor pyrite disseminations in the matrix. Alteration

minerals: clays, calcite, zeolites, pyrite.

364-408 m Trachyte: Grey, fine grained crystalline rock. It is phyric with sanidine and quartz

porphyries. Numerous specs of opaque minerals (olivine and pyroxenes) are noted as well

in the rock matrix. It depicts flow banding texture and slight intensity of alteration. Minor

pyrite is disseminated in the rock matrix. Alteration minerals: clays, zeolites, pyrite.

408-422 m Tuffs (lithic): Brown, highly vesicular tuffaceous rock. Vesicles are infilled with

chalcedony (412-414 m) and zeolites (scolecite and mesolite). The rock is weakly altered

with minor pyrite disseminations in the matrix. Alteration minerals: clays, calcite,

chalcedony, zeolites, pyrite.

422-476 m Trachyte: Grey, fine grained phyric rock with sanidine and quartz porphyries.

Specs of mafic minerals (olivine and pyroxenes) are noted in the rock matrix. It is slightly

94

vesicular, vesicles are infilled with calcite and chalcedony. Aveinlet infilled with chalcedony

noted between 446-448 m. Minor pyrite is disseminated in the rock matrix. Alteration

minerals: clays, calcite, titanite, chalcedony, pyrite.

476-498 m Tuff (lithic): Light grey to light brown, highly vesicular tuffaceous rock

intermixed with minor trachytic fragments. The latter are suspected to have fallen from the

unit above. Distinct vesicles infilled with secondary quartz (e.g. 476-478 m), zeolites (e.g.

480-482 m) and calcite are observed. Furthermore, wairakite is noted in a vesicle between

494-496 m.The rock is weakly altered with minor pyrite disseminations in the matrix.

Alteration minerals: clays, calcite, secondary quartz, wairakite, pyrite.

498-504 m Basalt: Dark grey, fine grained, slightly vesicular rock. Tuffaceous fragments

(suspected to have fallen from above) are notable in minor quantities. Quartz veins are

present between 500-502 m. It is slightly altered. Alteration minerals: clays, calcite,

secondary quartz.

504-564 m Trachyte: Grey, mixed cuttings of trachye (most abundant), tuffaceous and

basaltic fragments. It is veined and vesicular. Quartz vein noted between 516-518 m while

vesicles infilled with calcite are observed between 520-522 m. Alteration minerals: clays,

secondary quartz, calcite.

564-582 m Tuff (lithic): Mostly brown, whitish to brownish cryptocrystalline tuffaceous

fragments. It is mixed with pumiceous, trachytic and rhyolitic fragments and moderately

vesicular rock. Majority of the vesicles are empty, whereas, a few are infilled with zeolites

and secondary quartz. Pyrite is embedded as well as disseminated in the rock matrix. Minor

patches of green clays are noted replacing the matrix of individual fragments. Alteration

minerals: clays, calcite, pyrite, quartz, wairakite (574-578 m).

582-726 m Trachyte: Light grey, fine grained moderately porphyritic rock with feldspar and

quartz porphyries. Veinlets infilled with mafic minerals are discernible. Green clays are

observed replacing the mafic silicates. It is mildly oxidized and appears relatively fresh.

Partial loss between 596-604m and 636-638 m. Alteration minerals: clays, titanite.

726-744 m Tuff (lithic): Pale brown fine grained tuffaceous rock (dominant) mixed with

trachytic fragments. Occasional quartz phenocrysts are observed on both fragments. It is

moderately to highly oxidized but displays minimal intensity of alteration. Calcite is

observed infilling the vesicles. Alteration minerals: clays, calcite.

744-822 m Trachyte: Brown to dark grey, fine grained to cryptocrystalline rock. It is slightly

porphyritic with euhedral sanidine phenocrysts, some elongated in form. Mafic minerals are

outstanding in the matrix and show alteration to green clays. Albite noted replacing the

sanidine phenocrysts at 802-814 m and 816-818 m. It is mixed with minor tuffaceous

fragments. The formation shows minimal intensity of alteration. Alteration minerals: clays,

calcite, chalcedony, albite.

822-890 m Trachyte: Holocrystalline, grey to dark grey, fine grained rock depicting flow

banding texture (between 862-864 m). It is slightly porphyritic with subhedral to euhedral

phenocrysts of sanidine. Generally, the formation is massive, though vesicular between 870-

872 m, with green clays infilling the vesicles. Green clays are also observed extensively

replacing the mafic silicates. Albite is observed replacing sanidine phenocryst between 842-

844 m. The formation is slightly oxidized and exhibit slight degree of alteration. Alteration

minerals: clays, calcite, chalcedony, albite.

95

890-918 m Trachyte: Light grey to grey, fine grained poorly phyric rock with elongated

subhedral sanidine phenocrysts. It is relatively massive and holocrystalline. The rock is

slightly altered but shows high intensity of alteration between 904-910 m. Tuffaceous

fragments observed between 910-912 m (inferred to have fallen from above). Alteration

minerals: clays, calcite.

918-952 m Trachyte: Light grey, aphanitic rock. It is moderately feldspar phyric with the

dominant phenocryst phase being sanidine. Some phenocrysts are elongated and euhedral. It

is slightly altered. It is both slightly oxidized and altered. Alteration minerals: clays, calcite,

pyrite, albite.

952-966 m Tuff (lithic): Brown (952-962 m) to grey (962-966 m), fine grained tuffaceous

rock. It is highly vesicular between 962-966 m, vesicles are empty. It is moderately altered

to clays between 952-962 m and shows slight alteration for the remaining part. Alteration

minerals: clays, calcite, pyrite.

966-1046 m Trachyte: Grey, fine grained highly vesicular rock. It is moderately porphyritic

with subhedral sanidine phenocrysts. Most vesicles are empty and others partially infilled

with calcite. In addition, specs of mafic minerals are present, some being replaced by brown

and green clays. The formation is slightly altered but alteration intensity increases towards

the bottom (1024-1042 m). Partial loss between 1042-1046 m. Pyrite disseminated in the

rock matrix. Alteration minerals clays, pyrite, calcite, chalcedony (990-992 m).

1046-1150 m Rhyolite: Light grey to pink, fine grained massive and crystalline rock. It is

moderately porphyritic with quartz phenocrysts. Several specs of mafic minerals are noted

disseminated in the rock matrix. Some mafics are partially replaced by clays. The formation

is highly oxidized between 1054-1056 m and relatively fresh. Alteration minerals: clays,

pyrite.

1150-1198 m Basalt: Mostly brown, fine grained, mixed cuttings composed primarily of

basaltic fragments. It is slightly porphyritic with plagioclase phenocrysts. Majority of the

phenocryts have been replaced by clays whereas others are noted being replaced by calcite.

Mineral sequence noted between 1188-1190 m with chalcedony on the rim and calcite

precipitated on the core. Vesicles infilled with calcite are plentiful. Pyrite is disseminated in

the rock matrix. The formation is highly altered. Partial loss between 1152-1154 m and 1190-

1192 m. Alteration minerals: clays, pyrite, calcite, haematite (1192-1198 m).

1198-1214 m Basalt: Dark grey, fine grained to cryptocrystalline rock. It is poorly

porphyritic with indistinct plagioclase phenocrysts. Depicts slight intensity of alteration with

minor pyrite disseminations in the rock matrix. Alteration minerals: clays, pyrite, calcite,

haematite (1198-1200 m).

1214-1224 m Rhyolite: Light grey, fine grained highly felsic massive rock mixed with

cryptocrystalline basaltic fragments. It is phyric with quartz phenocrysts. Fresh specs of

mafic minerals are observed in the matrix. The formation is mostly fresh though slight

alteration to green and brown clays is evident. It is mildly oxidized as well. Alteration

minerals: clays.

1224-1260 m Trachyte: Light grey to pale green, fine grained holocrystalline rock but mostly

grey between 1234-1244 m. It is slightly porphyritic with sanidine phenocrysts. Minor

tuffaceous fragments noted between 1248-1260 m. The rock is slightly altered and

moderately oxidized between 1258-1260 m. Alteration minerals: titanite, clays, pyrite.

96

1260-1312 m Tuff (vitric): Mostly whitish, fine grained tuffaceous rock. It is highly vesicular

especially between 1294-1298 m, vesicles are infilled with calcite and green clays. Pyrite

dissemination (including euhedral cubes) is rather intense. This is a probable feed zone.

Alteration minerals: clays, calcite, pyrite.

1312-1440 m Trachyte: Light grey, to grey to pale green, fine grained rock. It is slightly

porphyritic with occasional subhedral sanidine phenocrysts. Calcite is noted replacing some

phenocrysts. Mafic silicates are being replaced by brown clays. Pyrite is noted infilling

veinlets as well as disseminated in the rock matrix. The formation is moderately altered.

Alteration minerals: pyrite, clays, titanite, calcite.

1440-1600 m Tuff (lithic): Grey to pale green, fine grained tuffaceous rock mixed with minor

trachytic fragments embedded within the lithic fragments. Vesicles infilled with calcite are

noted. It is slightly altered and pyrite disseminations are evident on the rock matrix. Partial

losses between 1454-1458 m,1462-1466 m and 1512-1518 m. Alteration minerals: pyrite,

clays, titanite, calcite.

1600-1642 m Tuff (lithic): Grey, fine grained tuffaceous rock. It is highly vesicular, vesicles

are infilled with clays and calcite. Also, veinlets are noted infilled with calcite. It shows

incipient alteration with minor pyrite disseminations in the rock matrix. Alteration minerals:

pyrite, clays, titanite, calcite.

1642-1654 m Basalt: Dark grey, fine grained crystalline rock. It is slightly porphyritic with

plagioclase phenocrysts. Occasionally, vesicles infilled with calcite are present. Also, calcite

is observed replacing some plagioclase phenocrysts. It is moderately altered and pyrite is

disseminated in the rock matrix. Alteration minerals: pyrite, clays, titanite, calcite.

1654-1700 m Basaltic trachyandesite: Brown, slightly porphyritic crystalline rock with

plagioclase phenocryts. The latter are noted being replaced by clays and calcite. Mafic

silicates display total alteration to clays. Platy calcite observed between 1646-1648 m and

1680-1682 m. Pyrite cubes are disseminated in the rock matrix. The formation is highly

altered. Alteration minerals: clays, calcite, pyrite, titanite.

1700-1730 m Trachydacite: Grey, fine grained to cryptocrystalline poorly porphyritic rock

with minor sanidine phenocrysts. Occasionally vesicles are present infilled with clays and

calcite. Epidote colouration is evident on a sanidine phenocryst between 1728-1730 m.

Pyrite is disseminated in the rock matrix. Partial circulation loss between 1722-1730 m. The

formation is moderately altered. Alteration minerals: clays, calcite, pyrite, titanite, epidote,

quartz.

1730-2052 m Trachyte: Grey to pale brown, fine grained slightly phyric rock with indistinct

sanidine phenocrysts. Vesicles infilled with calcite, green clays and secondary quartz are

noted. Veinlets infilled with calcite, mafic minerals and brown clays are noted as well.

Calcite replacing sanidine phenocryst between 1768-1770 m. Platy calcite noted between

1944-1946 m. Epidote coloration noted between 1704-1706 m, 1770-1774 m and 1930-1932

m. It is mildly oxidized and depicts moderate intensity of alteration. However, the formation

is highly altered (rated at a scale of 2.5) between 1788-1852 m. Minor pyrite disseminated

in the matrix. Partial losses between 1794-1796 m, 1798-1806 m, 1810-1816 m, 1820-1824

m, 1836-1838 m, 1842-1844 m, 1926-1930 m, 1954-1958 m, 1964-1972 m, and 2014-2018

m. A possible feed zone is suspected within this zone. Alteration minerals: pyrite, clays,

secondary quartz, titanite, calcite, chalcedony (2008-2010 m), epidote.

97

2052-2070 m Trachyte: Grey, fine grained to cryptocrystalline rock with minor sanidine

phenocrysts. It is relatively dense with specs of mafic minerals conspicous in the matrix,

some showing alteration to brown clays. Trace amounts of pyrite are disseminated in the

matrix. It is slightly altered. Alteration minerals: clays, calcite, pyrite.

2070-2186 m Trachyte: Grey to pale brown, fine grained slightly phyric rock with indistinct

sanidine phenocrysts. It is highly vesicular. Vesicles infilled with calcite and green clays are

noted. Veinlets are noted as well, infilled with calcite, mafic minerals and brown clays.

Chalcedony observed recrystallizing to secondary quartz between 2114 and 2116 m. It is

mildly oxidized and depicts moderate intensity of alteration. Partial loss between 2070-2076

m, 2104-2114 m, 2128-2130 m, 2142-2150 m, 2158-2162 m, 2170-2172 m, and 2184-2186

m. A possible feed zone within this zone. Alteration minerals: pyrite, clays, titanite, calcite.

2186-2210 m Trachydacite: Pale grey to pink, fine grained aphyric rock mixed with grey

trachytic fragments. The rock is rather massive and depicts slight intensity of alteration.

Minor pyrite is embedded in the rock matrix. Partial loss between 2186-2192 m and 2206-

2210 m. Alteration minerals: clays, calcite, pyrite.

2210-2326 m Trachyte: Mostly grey though showing distinctive gradation to brown between

2256-2260 m and 2302-2306 m. It is sparsely porphyritic with occassional sanidine

phenocrysts. Minor basaltic and tuffaceous fragments are noted. Vesicles and veinlets

infilled with brown clays are noted. Minor pyrite is disseminated in the rock matrix. Calcite

deposition is quite intense in this zone (scale 2.5), possibly reflecting a minor feed zone. It

is moderately to highly altered. Partial circulation returns experienced between 2214-2216

m, 2246-2248 m, 2266-2268 m, 2276-2278 m Alteration minerals: clays, haematite, pyrite,

calcite, titanite, epidote coloration (2234-2236 m).

2326-2370 m Trachydacite: Pink, highly felsic fine grained slightly porphyritic rock with

primarily quartz phenocrysts. Minor basaltic fragments are observed between 2334-2342 m.

Vesicles are noted infilled with secondary quartz and dark green clays. Veinlets are largely

infilled by brown clays and white minerals. The latter also replace the mafic silicates.

Euhedral pyrite cubes are disseminated in the matrix. It is moderately altered. Alteration

minerals: clays, pyrite, titanite, calcite, quartz.

2370-2592 m Rhyolite: Pale brown, fine grained, poorly porphyritic rock. It shows flow

texture and veinlets infilled with pyrite and brown minerals. Minor pyrite is disseminated in

the rock matrix. Generally, the formation is moderately to highly altered but shows slight to

moderate intensity of alteration from 2448 m to 2474 m. Partial loss of returns between 2374-

2380 m, 2388-2390 m, and 2398-2412 m. Alteration minerals: clays, pyrite, calcite,

haematite, chalcedony, epidote coloration (2570-2580 m).

2592-2780 m Trachydacite: Light grey to grey, fine grained slightly porphyritic rock with

subhedral to euhedral elongated sanidine phenocryts. It is moderately to strongly porphyritic

with striated sanidine phenocrysts between 2598-2610 m and 2618-2632 m. It is slightly

vesicular. Vesicles are infilled with brown clays and calcite. Pyrite is disseminated in the

rock matrix. The formation is moderately altered. Alteration minerals: clays, calcite, pyrite,

haematite,.

2780-2784 m Trachyandesite: Grey to dark grey, fine grained poorly porphyritic rock with

minor feldspar phenocrysts. It is relatively dense with numerous specs of mafic minerals

noted in the rock matrix. Epidote noted replacing feldspar phenocrysts (plagioclase) between

2782-2784 m. Paragenetic sequence, with epidote occuring in the core and calcite occuring

98

in the rim is also noted between the same depth interval. Minor pyrite disseminated in the

matrix. It shows slight degree of alteration. Alteration minerals: clays, calcite, pyrite, quartz,

epidote.

2784-2990 m Trachyte: Mostly grey, poorly porphyritic rock (though slightly porphyritic

between 2786-2794 m) with indistinct sanidine phenocrysts. It is slightly oxidized and both

vesicles and veinlets infilled with calcite and quartz are observed. Pyrite disseminations are

discernible in the matrix. Partial losses between 2816-2820 m. It is slightly altered.

Alteration minerals: clays, pyrite, haematite, titanite, calcite

99

Appendix B: XRD analysis for clay minerals in well OW-922.

OW-922-CLAY ANALYSIS RESULTS

Depth (m) d(001) A d(001) G d(001) H d(002) Mineral Type Other minerals

318-320 No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks

338-340 14.11 19.13 9.89 Smectite Sm. Feldspars, Amphiboles

358-360 12.48 16.77 9.97 Smectite Sm. Feldspars, Amphiboles

380-382 15.36 17.89 10.01 Smectite Sm. Feldspars

398-400 13.85 17.27 9.91 Smectite Sm. Feldspars, Amphiboles

418-420 No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks

438-440 15.66 18.05 10.11 Smectite Sm. Feldspars, Amphiboles

474-476 10.03 10.03 10.03 7.23, HIT=0 Illite + Chlorite (traces) Chl: ill Feldspars, Amphiboles

538-540 10.04 10.04 10.04 7.23, HIT=0 Illite + Chlorite (traces) Chl: ill Feldspars, Amphiboles

558-560 No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks

738-740 9.96 9.96 9.96 Illite (traces) ill Feldspars, Amphiboles

758-760 10.03 10.03 10.03 Illite (traces) ill Feldspars, Amphiboles

778-780 10.14 10.14 10.14 Illite (traces) ill Feldspars, Amphiboles

952-960 No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks

978-980 No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks

998-1000 No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks

1038-1040 15.41, 10.07 17.42, 10.07 10.07 7.18, HIT=0 Smectite+ Illite + Chlorite Sm:ill: Chl Feldspars, Amphiboles

1058-1060 13.69, 10.13 17.30, 10.13 10.13 7.31, HIT=0 Smectite + Illite Sm: ill Feldspars, Amphiboles

1198-1200 15.13 17.15 10.29 7.26, HIT=0 Smectite + Unst. Chlorite Sm: Chl Feldspars

1218-1220 12.92 17.24 9.86 7.07, HIT=0 Smectite + Unst. Chlorite Sm: Chl Feldspars

1340-1342 12.07 12.07 12.07 7.12, HIT=1 Unst. Chlorite + illite Chl: ill Feldspars

1358-1360 12.69, 10.06 17.09, 10.06 10.06 7.09, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles

100

1418-1420 12.86, 9.96 17.09, 9.96 9.96 7.13, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles

1438-1440 15.29, 9.99 17.17, 9.99 9.99 7.09, HIT=1 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles

1458-1460 15.34, 9.79 17.08, 9.79 9.79 7.08, HIT=2 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles

1554-1556 9.96 9.96 9.96 7.14, HIT=0 Chlorite + Illite Chl:ill Feldspars, Amphiboles

1558-1560 15.18, 10.11 17.26, 10.11 10.14 7.11, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles

1576-1578 12.89, 9.91 17.14, 9.91 9.91 7.06, HIT=1 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles

1578-1582 13.94, 9.93 16.99, 9.93 9.93 7.05, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles

1598-1600 12.69, 10.03 17.08, 10.03 10.03 7.04, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles

1618-1620 13.14, 9.84 16.99, 9.84 9.84 7.07, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles

1640-1642 15.36 17.17 9.91 7.17, HIT=0 Smectite+ Unst. Chlorite Sm:Chl. Feldspars, Amphiboles

1658-1660 14.02, 10.01 17.11, 10.01 10.01 7.09, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles

1696-1698 14.99, 9.98 16.9, 9.98 9.98 HIT=0 Smectite+Illite Sm:ill Feldspars

1698-1700 12.51, 10.19 17.17, 10.19 10.19 7.11, HIT=0 Smectite+ Illite+ Chlorite Sm:ill:Chl Feldspars

1716-1718 13.48, 10.07 17.57, 10.07 10.07 7.21, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

1718-1720 13.01, 9.89 16.99, 9.89 9.89 7.16, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

1738-1740 14.4, 9.98 16.99, 9.98 9.98 7.22, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

1740-1742 13.5, 9.96 17.26, 9.96 9.96 7.13, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

1760-1762 14.47, 10.11 17.26, 10.11 10.11 7.19, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

1778-1780 12.96, 9.99 17.38, 9.99 9.99 7.1, HIT=1 Smectite+ Illite+Chlorite Sm:ill:Chl

1780-1782 13.32, 9.96 17.17, 9.96 9.96 7.07, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

1800-1802 12.84, 9.91 16.54, 9.91 9.91 7.07, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

1818-1820 12.64, 9.98 17.20, 9.98 9.98 7.09, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

1838-1840 14.64, 9.89 16.63, 9.89 9.89 7.16, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl

1860-1862 15.23, 9.98 17.79, 9.98 9.98 7.15, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

1876-1880 14.68, 10.05 17.17, 10.05 10.05 7.08, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

1900-1902 14.19, 10.08 16.91, 10.08 10.08 7.08, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl

1918-1920 14.04, 9.92 14.04, 9.92 14.04, 9.92 7.06, HIT=0 Chlorite + Illite Chl: ill

1940-1942 14.53, 10.01 16.99, 10.01 10.01 7.13, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

1956-1958 14.26, 9.98 17.63, 9.98 9.98 7.15, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

1958-1960 9.98 9.98 9.98 7.15, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

1978-1980 9.84 9.84 9.84 7.12, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

1996-1998 10.03 10.03 10.03 7.09, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

101

2018-2020 9.96 9.96 9.96 7.14, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

2038-2040 13.11, 10.17 17.41, 10.17 10.17 7.19, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

2056-2058 10.12 10.12 10.12 7.15, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

2058-2060 13.16, 9.95 16.94, 9.95 9.95 7.05, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

2078-2080 14.06, 9.97 17.14, 9.97 9.97 7.04, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars

2096-2098 10.34 10.34 10.34 7.13, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

2118-2120 14.28, 9.82 14.28, 9.82 14.28, 9.82 7.06, HIT=0 Chlorite + Illite Chl: ill Feldspars

2156-2158 9.94 9.94 9.94 7.13, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

2196-2200 14.12, 9.96 17.12, 9.96 9.96 7.08, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl

2198-2200 14.59, 10.09 14.59, 10.09 14.59, 10.09 7.13, HIT=0 Chlorite + Illite Chl: ill Feldspars

2240-2242 12.56 17.12 9.73 7.03, HIT=0 Smectite+ Chlorite Sm:Chl

2258-2262 14.73, 10.2 14.73, 10.2 14.73, 10.2 7.26, HIT=0 Chlorite + Illite Chl: ill Feldspars

2278-2282 15.2, 10.31 15.2, 10.31 15.2, 10.31 7.18, HIT=0 Chlorite + Illite Chl: ill Feldspars

2296-2300 14.02, 9.94 16.97, 9.94 9.94 7.07, HIT=0 Smectite+ + Illite+Chlorite Sm:ill:Chl

2316-2320 14.06, 10.16 16.94, 10.16 10.16 7.04, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl

2338-2340 10.06 10.06 10.06 7.04, HIT=0 Unst. Chlorite + Illite Chl: ill

2338-2342 12.76, 9.89 16.94, 9.89 9.89 7.03, HIT=0 Smectite+ + Illite+Chlorite Sm:ill:Chl Feldspars

2354-2358 14.02, 10.11 17.26, 10.11 10.11 7.09, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl

2458-2460 14.14, 10.14 17.39, 10.14 10.14 7.09, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl

2482-2484 - - - 7.14, HIT=0 Unst. Chlorite Chl. Feldspars

2496-2500 10.01 10.01 10.01 7.07, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

2516-2518 9.99 9.99 9.99 7.09, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

2522-2524 10.86 10.86 10.86 7.12, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

2536-2540 12.55, 9.99 17.03, 9.99 9.99 7.08, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars

2558-2562 14.45, 9.89 17.03, 9.89 9.89 7.09, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl

2576-2580 13.01, 9.92 17.2, 9.92 9.92 7.11, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl

2616-2618 12.48, 10.15 17.54, 10.15 10.15 7.08, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl

2676-2678 14.77, 10.03 17.23, 10.03 10.03 7.08, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars

2696-2700 12.21, 9.91 17.14, 9.91 9.91 7.05, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl

2716-2718 12.03, 9.91 12.02, 9.91 12.02, 9.91 7.09, HIT=0 Chlorite + Illite Chl: ill Feldspars

2758-2760 10.65 10.65 10.65 7.07, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

2760-2762 10.72 10.72 10.72 7.09, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

102

2798-2800 9.98 9.98 9.98 7.12, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

2834-2836 14.36, 9.89 17.51, 9.89 9.89 7.13, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars

2840-2842 12.42, 10.06 17.15, 10.06 10.06 7.09, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars

2898-2900 13.16, 9.98 16.99, 9.98 9.98 7.04, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars

2918-2920 12.89, 10.14 17.54, 10.14 10.14 7.12, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars

2958-2960 10.05 10.05 10.05 7.07, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars

2978-2980 13.06, 9.95 17.26, 9.95 9.95 7.11, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars

103

Appendix C: ICP-OES analysis

procedure

The first step is to grind the drill chips in agate mortar to a fineness of approximately 100

mesh powder. Always remember to clean the mortar and pestle using aceton after

gringing each sample to eliminate contamination between the samples.

Weigh 100 mg ± 1mg of the sample in an epicure graphite crucible. Reset the scale to 0

g and add 200 mg ± 1 mg of flux (Lithium metaborate, LiBO2) to the sample. Mix the

two properly making sure that the mixture does not spill on the crucible walls. The

purpose of the flux is to increase the melting temperature of the sample.

Prepare the working standard samples. In this case, a total of six standard samples were

prepared. These include three external reproducibility standards (K-1919, BIR-1 and

BGM-1) and three internal calibration standards (A-THO, B-THO and B-ALK). 250 mg

± 1 mg of the sample and 500 mg ± 1 mg of the flux were used.

The samples were melted in an electric furnace heated at 1000oC for at least 30 mins and

thereafter left to cool for another 15-20 mins before the fused bead (or glass pellet) is

transferred into a bottle containing 30 ml of Rockan complexing acid. The latter, basically

is a mixture of de-ionized water, 5% HNO3, 1.33% HCL and 1.33% semi-saturated Oxalic

acid, H2C2O4. The three calibration standards were however, transferred into three

different bottles, each containing 75 ml of Rockan complexing acid. Highly acidic/highly

alkaline solutions are not recommended for this analysis because they will extinguish the

argon plasma.

The glass pellet and Rockan complexing acid are immediately run in a carousel to

complete solution to avoid silica precipitation. Complete solution is achieved between 2-

3 hours. The samples are now ready for analysis at the ICP-OES machine.

Analytical session began by running the three calibration standards. Spectra vision

software was applied in the calculation of two to three point calibration for every single

element. Instrumental reference sample (REF), used for monitoring drift during analysis

was made of equal parts (a third) of the internal calibration standards. The REF was

analysed at the beginning of the session and at 10 samples interval across the whole

analytical period. One calibration standard (B-THO) was analysed in between a batch of

10 samples to observe instrument reproducibility in multiple analyses.

During data processing, the raw data was copied and pasted into a correction spreadsheet.

Thereafter, all the batch analyses were normalized to 100%. The spreadsheet calculates

time dependent variation along each column of the analysis.

Application of the time-dependent variation-correction to the batch results to equal

absolute values of the reference samples, both at the beginning and the end.

105

Appendix D: EMP analysis for chlorite and epidote

1. QUANTITATIVE ANALYSIS RESULTS FOR CHLORITE

Representative EMP analyses of Chlorite-Well OW-917- 1870 m depth

Point 1 2 3 4 5 6 7 8 9 10

SiO2 22.78 22.62 22.26 19.58 24.59 22.93 22.67 22.62 22.67 20.09

TiO2 0.09 0.08 0.06 0.04 0.04 0.08 0.06 0.02 0.08 0.00

Al2O3 18.46 18.72 18.3 16.33 17.81 18.67 18.39 18.05 18.73 15.93

FeO 34.38 34.55 33.32 34.55 30.82 33.88 34.62 33.60 33.76 33.64

MnO 0.70 0.85 0.80 0.70 1.07 0.65 0.78 0.66 0.63 0.66

MgO 5.42 4.84 4.81 5.3 6.1 4.96 5.83 4.91 5.01 5.93

CaO 0.05 0.04 0.10 0.07 0.07 0.05 0.03 0.06 0.06 0.05

Na2O 0.42 0.33 0.44 0.47 1.07 0.42 0.39 0.39 0.29 0.33

K2O 0.58 0.39 0.56 0.58 0.65 0.61 0.47 0.53 0.50 0.62

SrO 0.00 0.00 0.02 0.00 0.00 0.04 0.00 0.00 0.12 0.06

Total 82.88 82.40 80.68 77.62 82.22 82.29 83.25 80.84 81.86 77.32

Number of cations on the basis of 24 Oxygens

Si 4.688 4.684 4.703 4.417 5.000 4.739 4.651 4.765 4.708 4.519

Ti 0.015 0.012 0.009 0.007 0.006 0.012 0.009 0.003 0.013 0.000

Al 4.480 4.569 4.558 4.342 4.267 4.548 4.447 4.481 4.586 4.225

Fe2+ 5.919 5.985 5.889 6.518 5.241 5.855 5.941 5.919 5.864 6.330

Mn 0.122 0.148 0.143 0.133 0.184 0.114 0.136 0.118 0.111 0.126

Mg 1.663 1.495 1.515 1.783 1.850 1.528 1.785 1.541 1.551 1.988

Ca 0.012 0.009 0.023 0.017 0.015 0.011 0.007 0.014 0.014 0.013

Na 0.166 0.131 0.182 0.207 0.424 0.168 0.156 0.161 0.119 0.146

K 0.152 0.103 0.150 0.167 0.170 0.160 0.122 0.142 0.133 0.178

Sr 0.000 0.000 0.003 0.000 0.000 0.005 0.000 0.000 0.015 0.008

Total 17.216 17.135 17.176 17.591 17.156 17.140 17.255 17.144 17.113 17.532

106

Representative EMP analyses of Chlorite-Well OW-922- 2236 m depth

Point 1 2 3 4 5 6 7 8 9 10

SiO2 22.20 20.04 20.23 21.01 18.88 19.86 21.38 21.52 22.54 19.71

TiO2 0.10 0.10 0.08 0.05 0.02 0.01 0.02 0.07 0.02 0.01

Al2O3 14.86 14.57 13.63 14.49 13.11 13.51 13.29 14.84 13.36 13.27

FeO 30.50 31.78 29.63 29.54 28.47 28.19 28.85 29.93 29.76 29.72

MnO 0.12 0.13 0.11 0.11 0.08 0.11 0.14 0.13 0.11 0.12

MgO 5.38 5.88 4.73 4.58 4.47 4.41 4.32 4.87 4.42 4.92

CaO 0.10 0.12 0.05 0.07 0.08 0.08 0.05 0.08 0.05 0.05

Na2O 0.75 0.51 0.32 0.32 0.30 0.32 0.27 0.33 0.33 0.29

K2O 0.32 0.31 0.31 0.32 0.38 0.45 0.29 0.48 0.27 0.34

SrO 0.03 0.00 0.00 0.00 0.00 0.05 0.00 0.00 0.00 0.00

Total 74.35 73.45 69.09 70.50 65.78 66.99 68.62 72.25 70.85 68.44

Number of cations on the basis of 24 Oxygens

Si 5.050 4.702 4.997 5.045 4.919 5.039 5.265 5.039 5.365 4.941

Ti 0.017 0.018 0.014 0.008 0.003 0.002 0.004 0.012 0.003 0.002

Al 3.985 4.031 3.968 4.102 4.026 4.041 3.857 4.096 3.748 3.920

Fe2+ 5.803 6.236 6.122 5.934 6.204 5.982 5.940 5.861 5.924 6.231

Mn 0.023 0.026 0.024 0.023 0.018 0.024 0.030 0.025 0.021 0.025

Mg 1.824 2.057 1.742 1.638 1.736 1.668 1.584 1.699 1.567 1.837

Ca 0.024 0.030 0.014 0.018 0.021 0.021 0.013 0.020 0.012 0.014

Na 0.329 0.233 0.151 0.151 0.151 0.159 0.127 0.152 0.153 0.143

K 0.092 0.094 0.098 0.099 0.125 0.146 0.093 0.144 0.082 0.109

Sr 0.004 0.000 0.000 0.000 0.000 0.007 0.000 0.000 0.000 0.000

Total 17.151 17.426 17.130 17.018 17.203 17.090 16.913 17.048 16.876 17.222

107

Representative EMP anayses of Chlorite-Well OW-922- 2782 m depth

Point 1 2 3 4 5 6 7 8 9

SiO2 23.42 32.83 23.41 22.12 20.96 24.07 30.84 27.84 20.75

TiO2 0.02 0.04 0.04 0.06 0.00 0.01 0.01 0.00 0.00

Al2O3 15.51 12.45 15.38 13.85 14.26 14.65 15.47 10.01 13.46

FeO 35.23 29.59 33.21 32.97 30.77 33.49 28.02 26.06 32.73

MnO 0.24 0.20 0.26 0.33 0.28 0.26 0.32 0.22 0.29

MgO 4.07 4.08 4.33 5.38 4.66 5.08 4.01 2.83 3.81

CaO 0.14 0.09 0.08 0.16 0.14 0.16 0.13 0.12 0.18

Na2O 0.09 0.14 0.33 0.18 0.17 0.08 1.69 0.17 0.12

K2O 0.13 0.11 0.12 0.13 0.18 0.14 0.12 0.19 0.12

SrO 0.07 0.00 0.00 0.00 0.00 0.03 0.00 0.00 0.08

Total 78.92 79.54 77.17 75.18 71.41 77.97 80.61 67.44 71.55

Number of cations on the basis of 24 Oxygens

Si 5.092 6.618 5.154 5.050 5.007 5.241 6.139 6.681 5.029

Ti 0.003 0.007 0.007 0.011 0.000 0.002 0.002 0.001 0.000

Al 3.975 2.960 3.991 3.727 4.015 3.760 3.629 2.832 3.844

Fe2+ 6.405 4.989 6.116 6.296 6.148 6.098 4.664 5.232 6.633

Mn 0.044 0.035 0.049 0.064 0.057 0.047 0.055 0.044 0.060

Mg 1.318 1.226 1.421 1.831 1.658 1.650 1.189 1.013 1.376

Ca 0.034 0.019 0.018 0.039 0.035 0.038 0.027 0.031 0.046

Na 0.038 0.055 0.143 0.080 0.078 0.035 0.651 0.078 0.058

K 0.035 0.029 0.033 0.039 0.053 0.038 0.030 0.059 0.039

Sr 0.009 0.000 0.000 0.000 0.000 0.004 0.000 0.000 0.012

Total 16.953 15.937 16.932 17.136 17.051 16.913 16.386 15.971 17.097

108

Representative EMP analyses of Chlorite-Well OW-910- 864 m depth

Point 1 2 3 4 5 6 7 8 9 10

SiO2 24.59 25.16 23.22 24.92 24.25 25.25 25.02 25.04 25.90 23.93

TiO2 0.03 0.02 0.04 0.03 0.05 0.00 0.03 0.00 0.01 0.04

Al2O3 13.23 13.51 12.57 12.86 11.94 13.14 12.95 13.68 13.34 12.34

FeO 12.16 12.24 12.67 12.46 12.02 12.20 12.60 11.96 12.33 12.03

MnO 0.21 0.26 0.20 0.19 0.18 0.21 0.19 0.23 0.19 0.21

MgO 17.05 16.63 17.30 16.72 16.21 16.70 17.58 15.63 15.73 17.10

CaO 0.29 0.28 0.27 0.37 0.32 0.31 0.31 0.29 0.30 0.33

Na2O 0.00 0.06 0.04 0.02 0.03 0.01 0.03 0.01 0.02 0.04

K2O 0.02 0.00 0.02 0.01 0.02 0.01 0.02 0.01 0.02 0.01

SrO 0.06 0.08 0.00 0.02 0.00 0.03 0.00 0.00 0.00 0.01

Total 67.64 68.24 66.33 67.60 65.02 67.85 68.73 66.84 67.83 66.04

Number of cations on the basis of 24 Oxygens

Si 5.428 5.496 5.276 5.509 5.573 5.542 5.446 5.559 5.668 5.425

Ti 0.005 0.003 0.007 0.006 0.009 0.000 0.006 0.000 0.002 0.006

Al 3.441 3.477 3.367 3.350 3.233 3.399 3.323 3.582 3.440 3.297

Fe2+ 2.245 2.235 2.408 2.304 2.310 2.238 2.294 2.222 2.257 2.281

Mn 0.039 0.048 0.038 0.035 0.035 0.039 0.035 0.042 0.035 0.040

Mg 5.609 5.415 5.861 5.511 5.553 5.462 5.703 5.174 5.133 5.777

Ca 0.069 0.067 0.065 0.088 0.078 0.072 0.073 0.068 0.070 0.081

Na 0.000 0.024 0.019 0.009 0.014 0.004 0.013 0.003 0.007 0.019

K 0.004 0.000 0.005 0.002 0.006 0.002 0.005 0.002 0.005 0.003

Sr 0.008 0.010 0.000 0.003 0.000 0.004 0.000 0.000 0.000 0.002

Total 16.85 16.77 17.05 16.82 16.81 16.76 16.90 16.65 16.62 16.93

109

Representative EMP analyses of Chlorite-Well OW-910- 1236 m depth

Point 1 2 3 4 5 6 7 8 9 10

SiO2 30.45 31.82 35.11 33.18 30.43 29.62 37.91 33.01 30.81 32.81

TiO2 0.11 0.03 0.06 0.03 0.00 0.03 0.00 0.02 0.02 0.01

Al2O3 16.90 19.12 14.33 17.43 16.52 16.50 12.86 16.46 18.40 17.96

FeO 17.62 17.25 16.94 17.42 17.50 18.37 17.56 16.25 17.61 18.29

MnO 0.28 0.26 0.25 0.25 0.27 0.27 0.29 0.25 0.31 0.20

MgO 15.59 18.07 16.05 14.52 18.64 17.54 15.99 16.23 19.82 16.19

CaO 0.44 0.39 0.94 0.85 0.56 0.40 3.37 1.45 0.32 0.44

Na2O 0.09 0.25 0.06 0.20 0.16 0.05 0.04 0.39 0.03 0.16

K2O 0.01 0.04 0.04 0.22 0.06 0.02 0.05 0.10 0.03 0.17

SrO 0.02 0.02 0.00 0.05 0.04 0.00 0.01 0.01 0.02 0.00

Total 81.50 87.25 83.78 84.15 84.18 82.80 88.08 84.16 87.35 86.23

Number of cations on the basis of 24 Oxygens

Si 5.634 5.463 6.239 5.902 5.469 5.439 6.452 5.863 5.316 5.717

Ti 0.015 0.004 0.008 0.004 0.000 0.004 0.000 0.002 0.002 0.002

Al 3.686 3.870 3.003 3.655 3.500 3.572 2.581 3.445 3.743 3.689

Fe2+ 2.726 2.477 2.519 2.592 2.631 2.822 2.499 2.413 2.542 2.666

Mn 0.043 0.038 0.038 0.037 0.041 0.042 0.042 0.038 0.045 0.029

Mg 4.298 4.626 4.252 3.85 4.993 4.803 4.058 4.297 5.096 4.206

Ca 0.087 0.072 0.179 0.162 0.109 0.078 0.614 0.276 0.058 0.082

Na 0.031 0.082 0.020 0.068 0.057 0.017 0.013 0.133 0.009 0.054

K 0.004 0.009 0.009 0.050 0.014 0.005 0.012 0.023 0.007 0.037

Sr 0.002 0.002 0.000 0.005 0.004 0.000 0.001 0.001 0.002 0.000

Total 16.53 16.64 16.27 16.33 16.82 16.78 16.27 16.49 16.82 16.48

110

Representative EMP analyses of Chlorite-Well OW-905A- 1560 m depth

Point 1 2 3 4 5 6 7 8 9 10

SiO2 25.68 25.61 24.72 26.12 25.19 24.12 24.99 26.53 24.96 25.54

TiO2 0.03 0.00 0.05 0.01 0.00 0.04 0.05 0.00 0.08 0.02

Al2O3 19.48 19.16 19.08 19.53 18.94 22.21 18.95 19.79 18.92 19.29

FeO 35.60 34.41 35.64 35.47 36.21 31.47 36.55 35.91 35.22 35.78

MnO 2.20 2.02 2.20 2.13 2.26 1.98 2.36 2.25 2.38 2.25

MgO 7.08 7.87 7.23 7.44 7.08 7.87 7.55 7.36 8.25 7.84

CaO 0.18 0.11 0.13 0.13 0.11 0.13 0.12 0.15 0.15 0.10

Na2O 0.09 0.02 0.03 0.08 0.07 0.21 0.03 0.09 0.02 0.00

K2O 0.02 0.02 0.02 0.03 0.01 0.04 0.02 0.03 0.03 0.03

SrO 0.02 0.02 0.00 0.00 0.07 0.06 0.00 0.08 0.00 0.00

Total 90.39 89.26 89.09 90.94 89.94 88.13 90.62 92.19 90.00 90.85

Number of cations on the basis of 24 Oxygens

Si 4.802 4.821 4.714 4.837 4.766 4.533 4.703 4.851 4.701 4.757

Ti 0.004 0.000 0.007 0.001 0.000 0.006 0.007 0.000 0.011 0.002

Al 4.294 4.250 4.288 4.264 4.224 4.920 4.203 4.264 4.200 4.235

Fe2+ 5.568 5.418 5.684 5.495 5.729 4.946 5.752 5.490 5.546 5.574

Mn 0.348 0.322 0.355 0.335 0.362 0.315 0.376 0.349 0.379 0.355

Mg 1.973 2.209 2.054 2.054 1.998 2.206 2.117 2.006 2.315 2.177

Ca 0.037 0.023 0.027 0.026 0.022 0.026 0.025 0.030 0.029 0.021

Na 0.032 0.009 0.010 0.028 0.027 0.077 0.012 0.031 0.008 0.000

K 0.005 0.005 0.004 0.008 0.002 0.009 0.004 0.006 0.006 0.008

Sr 0.002 0.003 0.000 0.000 0.008 0.007 0.000 0.009 0.000 0.000

Total 17.07 17.06 17.14 17.05 17.14 17.05 17.20 17.04 17.20 17.13

111

2. QUANTITATIVE ANALYSIS RESULTS FOR EPIDOTE

Representative EMP analyses of Epidote-Well OW-922- 2782 m depth

Point 1 2 3 4 5 6 7 8 9 10

SiO2 37.48 37.50 37.34 37.65 37.71 37.56 37.58 37.23 37.30 37.60

TiO2 0.00 0.03 0.00 0.07 0.01 0.00 0.03 0.04 0.09 0.04

Al2O3 24.62 24.71 24.60 24.40 24.71 24.43 24.48 24.43 24.59 24.29

Fe2O3 12.71 12.48 12.49 12.44 12.59 12.57 12.38 12.46 12.29 12.44

MnO 0.11 0.17 0.13 0.13 0.19 0.14 0.14 0.16 0.09 0.17

MgO 0.11 0.05 0.08 0.08 0.07 0.07 0.07 0.08 0.10 0.07

CaO 22.97 22.81 22.91 22.94 22.90 22.48 22.65 22.73 22.64 22.50

Na2O 0.02 0.03 0.00 0.04 0.06 0.03 0.03 0.04 0.06 0.07

K2O 0.04 0.04 0.05 0.04 0.05 0.05 0.04 0.03 0.05 0.06

SrO 0.36 0.32 0.51 0.44 0.41 0.46 0.50 0.46 0.27 0.55

Total 98.43 98.13 98.10 98.23 98.71 97.79 97.91 97.66 97.49 97.79

Number of cations on the basis of 24 Oxygens

Si 5.694 5.707 5.695 5.729 5.710 5.736 5.733 5.702 5.708 5.747

Ti 0.000 0.003 0.000 0.008 0.001 0.000 0.004 0.004 0.010 0.005

Al 4.409 4.432 4.422 4.377 4.411 4.399 4.402 4.410 4.436 4.376

Fe 3+ 1.454 1.429 1.433 1.423 1.435 1.444 1.422 1.436 1.416 1.430

Mn 0.015 0.021 0.016 0.017 0.024 0.018 0.018 0.021 0.012 0.023

Mg 0.024 0.012 0.018 0.018 0.016 0.016 0.016 0.017 0.024 0.016

Ca 3.740 3.720 3.743 3.740 3.716 3.679 3.703 3.730 3.713 3.685

Na 0.006 0.010 0.000 0.013 0.018 0.010 0.010 0.013 0.017 0.020

K 0.008 0.007 0.009 0.008 0.010 0.010 0.009 0.006 0.009 0.011

Sr 0.032 0.028 0.045 0.039 0.036 0.040 0.044 0.041 0.024 0.049

Total 15.382 15.370 15.380 15.372 15.378 15.352 15.360 15.380 15.369 15.361

112

Representative EMP analyses of Epidote-Well OW-910- 864 m depth

Point 1 2 3 4 5 6 7 8 9 10

SiO2 38.00 37.67 37.70 37.80 37.74 37.92 37.32 38.14 37.49 37.98

TiO2 0.05 0.09 0.01 0.01 0.06 0.11 0.24 0.02 0.11 0.11

Al2O3 24.18 22.99 23.45 23.92 23.93 24.54 22.21 25.57 20.46 25.29

Fe2O3 13.37 15.02 14.38 13.90 13.68 13.28 16.14 11.86 18.56 12.01

MnO 0.02 0.09 0.00 0.06 0.10 0.08 0.14 0.04 0.09 0.06

MgO 0.13 0.13 0.06 0.08 0.06 0.06 0.09 0.06 0.11 0.06

CaO 23.23 23.16 23.38 23.33 23.62 23.38 23.21 23.15 22.80 23.38

Na2O 0.00 0.06 0.00 0.01 0.00 0.01 0.00 0.00 0.00 0.02

K2O 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.01

SrO 0.19 0.00 0.16 0.09 0.21 0.27 0.10 0.22 0.00 0.22

Total 99.17 99.22 99.14 99.21 99.41 99.64 99.46 99.08 99.63 99.14

Number of cations on the basis of 24 Oxygens

Si 5.731 5.714 5.714 5.712 5.699 5.697 5.681 5.721 5.733 5.707

Ti 0.005 0.011 0.001 0.001 0.007 0.012 0.028 0.002 0.013 0.013

Al 4.299 4.109 4.190 4.260 4.259 4.345 3.984 4.521 3.688 4.478

Fe 3+ 1.517 1.714 1.640 1.581 1.554 1.501 1.849 1.339 2.136 1.358

Mn 0.002 0.012 0.000 0.008 0.012 0.010 0.018 0.005 0.012 0.008

Mg 0.030 0.030 0.014 0.019 0.014 0.013 0.021 0.014 0.025 0.013

Ca 3.753 3.764 3.797 3.778 3.822 3.764 3.785 3.721 3.736 3.763

Na 0.000 0.018 0.000 0.003 0.000 0.002 0.000 0.001 0.000 0.005

K 0.002 0.000 0.000 0.000 0.000 0.000 0.001 0.002 0.000 0.002

Sr 0.016 0.000 0.014 0.008 0.019 0.023 0.009 0.020 0.000 0.019

Total 15.355 15.372 15.370 15.370 15.386 15.368 15.375 15.346 15.343 15.366

113

Representative EMP analyses of Epidote-Well OW-910- 1236 m depth

Point 1 2 3 4 5 6 7 8 9 10

SiO2 37.64 35.24 36.30 39.07 46.13 44.57 36.16 35.77 35.48 35.81

TiO2 0.58 0.57 0.67 0.44 0.49 0.56 0.47 0.49 0.39 0.34

Al2O3 8.73 7.67 7.75 9.15 9.18 6.66 8.03 7.23 6.92 7.19

Fe2O3 18.66 20.21 20.37 19.53 14.88 21.70 19.91 21.11 21.34 22.18

MnO 0.49 0.51 0.44 0.28 0.26 0.42 0.48 0.39 0.36 0.30

MgO 0.50 0.62 0.41 0.40 0.60 0.36 0.35 0.29 0.32 0.49

CaO 30.05 32.50 32.36 29.69 22.75 24.41 32.59 33.05 33.16 32.44

Na2O 0.88 0.03 0.21 1.07 0.17 0.06 0.02 0.01 0.04 0.06

K2O 0.17 0.07 0.07 0.07 3.59 0.62 0.03 0.05 0.09 0.04

SrO 0.02 0.04 0.00 0.05 0.00 0.01 0.00 0.01 0.10 0.05

Total 97.71 97.47 98.58 99.74 98.05 99.37 98.04 98.41 98.19 98.90

Number of cations on the basis of 24 Oxygens

Si 6.146 5.873 5.958 6.219 7.191 6.957 5.961 5.916 5.900 5.897

Ti 0.071 0.072 0.083 0.053 0.058 0.066 0.058 0.061 0.048 0.042

Al 1.681 1.506 1.499 1.716 1.686 1.225 1.559 1.409 1.356 1.395

Fe 3+ 2.293 2.536 2.517 2.339 1.745 2.549 2.471 2.628 2.670 2.750

Mn 0.068 0.072 0.061 0.038 0.034 0.055 0.067 0.055 0.051 0.041

Mg 0.121 0.155 0.099 0.095 0.140 0.084 0.086 0.070 0.078 0.120

Ca 5.258 5.804 5.692 5.064 3.801 4.082 5.756 5.857 5.909 5.725

Na 0.279 0.010 0.067 0.329 0.051 0.018 0.005 0.004 0.013 0.020

K 0.035 0.015 0.015 0.013 0.713 0.123 0.006 0.010 0.019 0.008

Sr 0.002 0.004 0.000 0.004 0.000 0.001 0.000 0.001 0.009 0.005

Total 15.953 16.047 15.992 15.871 15.418 15.160 15.970 16.012 16.054 16.003

114

Representative EMP analyses of Epidote-Well OW-905A- 1560 m depth

Point 1 2 3 4 5 6 7 8 9 10

SiO2 37.12 36.39 36.95 38.03 37.29 37.88 34.16 37.32 38.01 37.42

TiO2 0.14 0.08 0.23 0.11 0.08 0.11 0.34 0.13 0.16 0.11

Al2O3 23.10 23.48 19.55 24.18 24.02 24.34 20.39 24.09 24.44 24.18

Fe2O3 14.86 13.78 19.53 13.82 13.30 13.73 15.45 13.41 13.75 13.15

MnO 1.53 1.05 0.94 0.93 1.03 1.01 0.62 1.46 1.61 0.69

MgO 0.05 0.02 0.02 0.02 0.01 0.03 0.03 0.03 0.01 0.00

CaO 21.50 22.16 21.76 22.39 21.79 21.90 21.88 21.51 21.69 22.42

Na2O 0.00 0.01 0.01 0.02 0.01 0.00 0.00 0.00 0.00 0.00

K2O 0.01 0.01 0.00 0.01 0.03 0.03 0.05 0.03 0.01 0.02

SrO 0.05 0.01 0.06 0.04 0.00 0.03 0.09 0.00 0.02 0.09

Total 98.35 97.00 99.05 99.54 97.55 99.07 93.01 97.98 99.71 98.08

Number of cations on the basis of 24 Oxygens

Si 5.693 5.647 5.720 5.725 5.721 5.722 5.599 5.709 5.715 5.711

Ti 0.016 0.010 0.027 0.012 0.009 0.012 0.042 0.015 0.019 0.012

Al 4.175 4.295 3.566 4.291 4.343 4.334 3.940 4.342 4.331 4.349

Fe 3+ 1.715 1.609 2.275 1.566 1.536 1.561 1.906 1.544 1.555 1.510

Mn 0.199 0.139 0.124 0.118 0.133 0.129 0.086 0.189 0.205 0.090

Mg 0.011 0.005 0.004 0.004 0.003 0.007 0.008 0.007 0.003 0.000

Ca 3.534 3.684 3.610 3.612 3.581 3.546 3.842 3.525 3.495 3.666

Na 0.000 0.002 0.002 0.005 0.002 0.001 0.000 0.000 0.000 0.000

K 0.001 0.002 0.000 0.002 0.005 0.006 0.010 0.007 0.001 0.004

Sr 0.004 0.001 0.005 0.003 0.000 0.003 0.009 0.000 0.002 0.008

Total 15.348 15.393 15.333 15.338 15.334 15.321 15.442 15.337 15.326 15.350